robert@gar.no writes:
>In <199707291709.KAA10784@nlanr.net> wessels@nlanr.net (Duane Wessels) writes:
>
>>Try this patch included below, and add this to your squid.conf:
>> reload_into_ims on
>>This patch is designed to have the following effects
>> * if the request includes both 'no-cache' and 'If-Modified-Since'
>> the 'no-cache' is simply ignored
>> * if the request includes only 'no-cache' then Squid
>> makes an 'If-Modified-Since' request out the other side.
>
>Thanks, that mostly did the job. However I think there is a bug in the
>handling of the 'request includes both no-cache and If-Modified-Since' case.
>This should not blindly be turned into an IMS request without first
>checking the refresh_patterns for staleness. The Netscape simple reload
>problem reported by Christian Balzer is a side-effect of this I think.
>I've applied the patch below on top of yours with good results.
I'm not really convinced. And I like how you call this a bug
('scuse me! you're not violating the protocol properly!)
With the original patch I sent a NOCACHE+IMS request is not forwarded,
and you don't really know if the page is stale.
With your patch, such a request would be forwarded. If you have a
parent cache, then the parent may or may not forward it on to the
origin (you may or may not see stale data). If you have no parents,
then the origin server will tell you if you have the current version
(you should never see stale data).
For some reason I thought the goal was to NOT forward the request
because the request forwarding caused delays. But maybe everyone has
something different in mind.
Maybe it would be better to just leave NOCACHE+IMS requests
alone (so they always go all the way to the origin) and
only munge NOCACHE requests without IMS headers.
Duane W.
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id QAA10109;
Thu, 7 Aug 1997 16:26:41 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 16:26:41 -0700 (PDT)
Date: Fri, 08 Aug 1997 09:11:31 +1000 (EST)
From: David Richards <dj.richards@qut.edu.au>
Subject: Re: log files
In-reply-to: <33EA23A2.6E9B@cruiser.com.br>
To: "Vinicius D 'Antonio" <vda@cruiser.com.br>
Cc: user-list-squid <squid-users@nlanr.net>
Message-id: <Pine.OSF.3.92.970808090814.11844A-100000@pigeon.qut.edu.au>
MIME-version: 1.0
Content-type: TEXT/PLAIN; charset=US-ASCII
Resent-Message-ID: <"_qgDv20XJLB.0.zN7.oZbwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/124
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
Things to check:
1) The log directory exists
2) Squid owns the directory. i.e. the user that squid is running
as.
If I was running squid as say user "proxy" then got to squid's
home directory something like "/usr/local/squid" and type:
# chown -R proxy.proxygroup *
Assuming squid is running as "proxy" and is a member of
"proxygroup"
Sorry about being so specific.
Dave.
----------------------------------------------------------------------
David Richards Ph: +61 7 3864 4354
Network Programmer Fax: +61 7 3864 5272
Computing Services e-mail: dj.richards@qut.edu.au
Queensland University of Technology
----------------------------------------------------------------------
On Thu, 7 Aug 1997, Vinicius D 'Antonio wrote:
> Hi all,
>
> I've a problem! I can't start the squid, it's sending me a msg saying:
> WARNING: Cannot write log file: /usr/local/squid/logs/cache.log
> /usr/local/squid/logs/cache.log: Permission denied
> messages will be sent to 'stderr'.
>
> and
>
> file_open: error opening file /usr/local/squid/logs/access.lo
> g: (13) Permission denied
> 96/08/07 16:35:25| /usr/local/squid/logs/access.log: (13) Permission
> denied
> FATAL: Cannot open logfile.
>
> thanks in advance.
>
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id QAA10482;
Thu, 7 Aug 1997 16:48:44 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 16:48:44 -0700 (PDT)
Message-ID: <199708072346.JAA12205@diji.jna.com.au>
From: John Warburton <JWar@jna.com.au>
To: Oskar Pearson <oskar@is.co.za>,
Andre Albsmeier
<Andre.Albsmeier@mchp.siemens.de>
Cc: squid-users@nlanr.net
Subject: RE: get "Operation already in progress" when trying to use squid
with socks
Date: Fri, 8 Aug 1997 09:50:15 +1000
X-Priority: 3
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.0.1457.3)
Content-Type: multipart/mixed;
boundary="---- =_NextPart_000_01BCA3E0.296DBA30"
Resent-Message-ID: <"ua4r6-YNfh3.0.P_2.uvbwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/125
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.
------ =_NextPart_000_01BCA3E0.296DBA30
Content-Type: text/plain
Hi,
we use socks 5 with our squid with no problems.
> > The socks5 package itself runs great but with squid I had no succes.
> > > I socksified squid using the usual procedure: including socks.h,
> > > defining SOCKS and linking with the socks5 library.
> you no longer have to go through that rigmarole any more. We use the
> "runsocks" shell script which sets up the correct socks environment,
> so you don't have to modify the source. Our RunCache script has been
> modified. Instead of
> squid -sY $conf >> $logdir/squid.out 2>&1
> we have
> /usr/local/bin/runsocks squid -sY $conf >> $logdir/squid.out
> 2>&1
>
> pretty simple eh?
>
>
> > > Do you have an idea or any hint what I might be doing wrong?
> > hmm - this error message means that your server is trying to connect
> > directly to the remote server (ie not via socks). There is probably
> something
> > wrong with the https source in squid so that it can't talk socks.
>
> we do (still) have an issue with ftpget failing with socks5 and
> solaris 2.5.1, giving us:
> Connection Failed
>
> The system returned:
>
> (125) Address already in use
>
> This means that:
>
> The remote site or server may be down. Please try again soon.
>
> which when we use an vanilla socks ftp client, we get through. But
> that's another story....
>
> John
------ =_NextPart_000_01BCA3E0.296DBA30
Content-Type: application/ms-tnef
Content-Transfer-Encoding: base64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------ =_NextPart_000_01BCA3E0.296DBA30--
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id RAA15273;
Thu, 7 Aug 1997 17:58:03 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 17:58:03 -0700 (PDT)
Date: Fri, 8 Aug 1997 02:54:21 +0200
Message-Id: <mail_out.33ea6e3d.robert@gmail.gar.no>
In-Reply-To: <199708072021.NAA14922@nlanr.net>
From: Robert Andersson <robert@gar.no>
To: Duane Wessels <wessels@nlanr.net>
Cc: squid-users@nlanr.net
Subject: Re: Netscape 4 ?
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
X-Gmail-MailClass: 1; infodate=33ea2e4d
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by nlanr.net id RAA20905
Resent-Message-ID: <"4csbhRb8mdD.0.Oc.qvcwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/126
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
Reply to the message of Thursday August 7, 1997 22:21 +0200
-----------------------------------------------------------------
> For some reason I thought the goal was to NOT forward the request
> because the request forwarding caused delays. But maybe everyone has
> something different in mind.
Yes, it may well be that we have slightly different ideas about what
the reload_into_ims flag should mean. The way I expected it to be:
o NOCACHE+IMS requests are handled *exactly* like regular IMS. In
other words we first go through the refresh_pattern checking and
only if it says the object is valid is the IMS taken into account.
o NOCACHE requests are handled as if the object is stale and needs
revalidation.
It seems we agree on the NOCACHE case. What I don't understand is why
you want NOCACHE+IMS to always regard the cached object as valid and
thus bypass the refresh_pattern staleness checking. At least to me it
doesn't feel right that NOCACHE+IMS is less inclined to forward the
request than just a plain IMS
> Maybe it would be better to just leave NOCACHE+IMS requests
> alone (so they always go all the way to the origin) and
> only munge NOCACHE requests without IMS headers.
At least on our site I see that the majority of the NOCACHE requests
coming in from the broken Netscape 4 browsers also have IMS set. Which
makes sense, the browsers have their own disk and/or memory cache and
generate IMS relative to it.
Regards, Robert.
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id TAA25522;
Thu, 7 Aug 1997 19:14:41 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 19:14:41 -0700 (PDT)
Date: Fri, 08 Aug 1997 10:15:15 +1000
From: Adam Rigby <a.rigby@qut.edu.au>
Subject: Cachemgr.cgi configuration
X-Sender: rigbya@pop.qut.edu.au
To: squid-users@nlanr.net
Message-id: <3.0.32.19970808101511.00556480@pop.qut.edu.au>
MIME-version: 1.0
X-Mailer: Windows Eudora Pro Version 3.0 (32)
Content-type: text/plain; charset="us-ascii"
Resent-Message-ID: <"0691n2WtXZC.0.fl4.9-dwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/127
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
Hi all,
I am running Digital Unix 4.0a, Squid 1.1.11, Apache 1.2 on the same box.
All are working well together.
The problem:
After starting cachemgr.cgi and selecting any option from the drop down
list the following message appears eg.
info:proxy.is.here:3128
dated Fri Aug 8 10:56:42 1997
ERROR
Access Denied
Sorry, you are not currently allowed to request:
cache_object://proxy.is.here/info@
from this cache, Please check with the cache administrator if you believe
this is incorrect.
<<End of Problem
Please help, this is really starting to annoy me!!
Thanks
~Adam
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id VAA24825;
Thu, 7 Aug 1997 21:54:52 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 21:54:52 -0700 (PDT)
From: Strange Alex <strange@mix.unicon.ru>
Message-Id: <199708080449.LAA03388@mix.unicon.ru>
Subject: Logs rotating...
To: squid-users@nlanr.net
Date: Fri, 8 Aug 1997 11:49:11 +0700 (NSD)
Reply-To: strange@mix.unicon.ru
X-Class: Fast
X-Mailer: ELM [version 2.4ME+ PL22 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Resent-Message-ID: <"rpOO7uRNGk.0.lK6.CMgwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/128
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
Hi 2 all !
Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ?
My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid`
nothing happend with my log files, even when I do so more than 10 times ;(
And please, tell my the proper way, how to rotate all squid log files .
WBR Strange Alex
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id WAA21496;
Thu, 7 Aug 1997 22:13:48 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 22:13:48 -0700 (PDT)
Date: Fri, 8 Aug 1997 15:08:01 +1000 (EST)
From: David J N Begley <david@avarice.nepean.uws.edu.au>
Reply-To: David J N Begley <d.begley@nepean.uws.edu.au>
To: Squid Users <squid-users@nlanr.net>
Subject: Re: Logs rotating...
In-Reply-To: <199708080449.LAA03388@mix.unicon.ru>
Message-ID: <Pine.LNX.3.96.970808150651.20772C-100000@avarice.nepean.uws.edu.au>
X-No-Archive: Yes
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Resent-Message-ID: <"a2i0adduZj2.0.nS5.Sfgwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/129
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
On Fri, 8 Aug 1997, Strange Alex wrote:
> Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ?
> My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid`
> nothing happend with my log files, even when I do so more than 10 times ;(
> And please, tell my the proper way, how to rotate all squid log files .
Ensure that the userid executing the kill -USR1 has access to send signals
to the Squid process (particularly true under Solaris 2 I've noticed, less
necessary under Solaris 1).
Cheers..
dave
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id WAA24055;
Thu, 7 Aug 1997 22:37:00 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 22:37:00 -0700 (PDT)
From: Jaeyeon Jung <jyjung@cosmos.kaist.ac.kr>
Message-Id: <199708080508.OAA02671@cosmos.kaist.ac.kr>
Subject: Re: Logs rotating...
To: strange@mix.unicon.ru
Date: Fri, 8 Aug 1997 14:08:29 +0900 (KST)
Cc: squid-users@nlanr.net
In-Reply-To: <199708080449.LAA03388@mix.unicon.ru> from "Strange Alex" at Aug 8, 97 11:49:11 am
X-Mailer: ELM [version 2.4 PL21-h4]
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-2022-kr
Content-Transfer-Encoding: 7bit
Resent-Message-ID: <"t0x8hfk5LE7.0.M72.5_gwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/130
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
>
> Hi 2 all !
>
> Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ?
> My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid`
> nothing happend with my log files, even when I do so more than 10 times ;(
wait a few more minutes, then you can see *.log.0
jyjung@cosmos.kaist.ac.kr
> And please, tell my the proper way, how to rotate all squid log files .
>
> WBR Strange Alex
>
>
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id XAA22797;
Thu, 7 Aug 1997 23:05:39 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 23:05:39 -0700 (PDT)
Date: Thu, 7 Aug 1997 23:56:30 -0600 (MDT)
From: David Hughes <davidh@csn.net>
Subject: Re: Most Stable Squid
To: Edward Henigin <ed@texas.net>
cc: Paul Black <squid@f18.dot.net.au>, squid-users@nlanr.net, noc@dot.net.au
In-Reply-To: <19970807133824.52715@texas.net>
Message-ID: <Pine.3.89.9708072301.A11663-0100000@teal.csn.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Resent-Message-ID: <"ZtMUAEGQ7jC.0.1-5.VPhwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/131
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
What is the most stable current version of Squid?
tx, david
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id XAA28870;
Thu, 7 Aug 1997 23:13:17 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 23:13:17 -0700 (PDT)
From: Strange Alex <strange@mix.unicon.ru>
Message-Id: <199708080608.NAA03519@mix.unicon.ru>
Subject: Re: Logs rotating...
To: squid-users@nlanr.net
Date: Fri, 8 Aug 1997 13:08:40 +0700 (NSD)
In-Reply-To: <Pine.LNX.3.96.970808150651.20772C-100000@avarice.nepean.uws.edu.au> from "David J N Begley" at "Aug 8, 97 03:08:01 pm"
Reply-To: strange@mix.unicon.ru
X-Class: Fast
X-Mailer: ELM [version 2.4ME+ PL22 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Resent-Message-ID: <"ivhzUYRP506.0.pp5.IXhwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/132
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
One day David J N Begley wrote:
> > Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ?
> > My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid`
> > nothing happend with my log files, even when I do so more than 10 times ;(
> > And please, tell my the proper way, how to rotate all squid log files .
>
> Ensure that the userid executing the kill -USR1 has access to send signals
> to the Squid process (particularly true under Solaris 2 I've noticed, less
> necessary under Solaris 1).
Root id is enough ??? ;)
And this is cache.log:
97/08/08 11:00:22| rotate_logs: SIGUSR1 received.
97/08/08 11:00:23| storeWriteCleanLog: Starting...
97/08/08 11:00:23| Finished. Wrote 1712 lines.
97/08/08 11:00:23| Took 0 seconds (1712.0 lines/sec).
97/08/08 11:00:23| storeRotateLog: Rotating.
97/08/08 11:00:23| stat_rotate_log: Rotating
All Ok, but really nothinh happend ! ;(((
And Jaeyeon Jung wrote:
> nothing happend with my log files, even when I do so more than 10 times ;(
wait a few more minutes, then you can see *.log.0
I wait more than 1 day ;) It should be quite enough ;)
WBR Strange Alex
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id XAA21749;
Thu, 7 Aug 1997 23:19:47 -0700 (PDT)
Resent-Date: Thu, 7 Aug 1997 23:19:47 -0700 (PDT)
Date: Fri, 8 Aug 1997 08:16:11 +0200 (CEST)
From: Daniel Schneider <daniel@terrania.westfalen.de>
To: Ed Sawkins <ed@renishaw.co.uk>
cc: squid-users@nlanr.net
Subject: Re: Starting Squid On Bootup
In-Reply-To: <33E9F04B.54F8@renishaw.com>
Message-ID: <Pine.LNX.3.95.970808081422.26383B-100000@terrania.westfalen.de>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by nlanr.net id XAA11705
Resent-Message-ID: <"rHPFvL4rHp8.0.nJ5.Idhwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/133
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
Hi,
On Thu, 7 Aug 1997, Ed Sawkins wrote:
> If I try to start with the normal rc scripts it just stops after
> 'testing dns'
Well, that sounds like the dns-test doesn't work quite right or does take
much time.
Just change the option in your squid.conf: dns_testnames
Daniel Schneider
"Das derzeitige Hochwasser ist in Ostdeutschland die
größte Katastrophe seit der Wiedervereinigung."
- Bundeskanzler H. Kohl
Received: (from list@localhost)
by nlanr.net (8.8.6/8.8.6) id BAA30744;
Fri, 8 Aug 1997 01:26:46 -0700 (PDT)
Resent-Date: Fri, 8 Aug 1997 01:26:46 -0700 (PDT)
Message-Id: <m0wwkEn-000772C@dingo.theplanet.co.uk>
X-Mailer: exmh version 2.0delta 6/3/97
To: ed@renishaw.co.uk
cc: squid-users@nlanr.net
From: Nigel Metheringham <Nigel.Metheringham@theplanet.net>
Subject: Re: Starting Squid On Bootup
In-reply-to: Your message of "Thu, 07 Aug 1997 16:56:59 -0000."
<33E9F04B.54F8@renishaw.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Fri, 08 Aug 1997 09:17:33 +0100
Sender: nigel@theplanet.net
Resent-Message-ID: <"VrZmWdCU4u4.0.oE.-Ujwp"@oceana.nlanr.net>
Resent-From: squid-users@nlanr.net
X-Mailing-List: <squid-users@nlanr.net> archive/latest/134
X-Loop: squid-users@nlanr.net
Precedence: list
Resent-Sender: squid-users-request@nlanr.net
} I have Squid running on a hp-ux 10.10 machine, and it works fine
} when started from the command line.
}
} If I try to start with the normal rc scripts it just stops after
} 'testing dns'
Your boot scripts *have* started the named prior to squid starting up??
If you primary/secondary a lot of domains there may be a significant delay
before named is able to fulfil requests.
[this applies to other things that squid depends on - ie basic networking,
name service (even disk mounting....!). Squid needs to be started
pretty late in the sequence]
Nigel.
-- [ Nigel.Metheringham@theplanet.net - Systems Software Engineer ] [ Tel : +44 113 251 6012 Fax : +44 113 224 0003 ] [ Friends don't let friends use sendmail! ] Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id BAA30537; Fri, 8 Aug 1997 01:39:07 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 01:39:07 -0700 (PDT) Message-ID: <33EADBD1.C6E66624@siemens.nl> Date: Fri, 08 Aug 1997 10:41:54 +0200 From: Gert-Jan Looy <gj@siemens.nl> Reply-To: gj@siemens.nl Organization: Siemens X-Mailer: Mozilla 4.0 [en] (WinNT; I) MIME-Version: 1.0 To: squid-users@nlanr.net Subject: ipcacheChangeKey... X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"v70U19rBVg3.0.Et1.Vijwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/135 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, After updating to Squid 1.1.14 on a SNI SVR4 box, strange messages keep popping up in cache.log: e.g.: .... ipcacheChangeKey: from 'www.microsoft.com' to '23/www.microsoft.com' .... ipcacheChangeKey: from 'www.microsoft.com' to '24/www.microsoft.com' .... ipcacheChangeKey: from 'www.microsoft.com' to '25/www.microsoft.com' etc. Is this an error? Everything seems to work fine though. TIA, _________________________________________________________ Gert-Jan Looy Unix SysAdm Siemens/SNI Nederland N.V. E-mail : gj@siemens.nl Phone/Fax : +31 70 333 2982 / 2948 X400 : NL;400NET;SCN;siemens;zom1;p1;looy;gert-jan Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id BAA10103; Fri, 8 Aug 1997 01:40:01 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 01:40:01 -0700 (PDT) From: Michael Shoyher <msh@corbina.net> Message-Id: <199708080839.MAA07986@ns.corbina.ru> Subject: FD_SETSIZE and FreeBSD 2.1.7 To: squid-users@nlanr.net Date: Fri, 8 Aug 1997 12:39:44 +0400 (MSD) X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"ArZ8wzFGlg3.0.217.Djjwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/136 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Is there any way to use more than 256 filedescriptors on FreeBSD 2.1.7? Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA28321; Fri, 8 Aug 1997 02:22:32 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 02:22:32 -0700 (PDT) Message-Id: <199708080925.LAA20149@chappi.deh.de> From: "Thomas Krause" <tom@deh.de> To: <squid-users@nlanr.net> Subject: cache_swap > 6 GB Date: Fri, 8 Aug 1997 11:23:29 +0200 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1161 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"LOwVcBqdTo6.0.2l6.wKkwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/137 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net I want to use a 7.3 GB cache partition for squid 1.1.14 If the cache_swap value is larger then 6000000 (eg. 6300000) I get the following message: WARNING: cache_swap (-2138734592 kbytes) is less than cache_mem (67108864 bytes). This will cause serious problems with your cache!!! Change your configuration file. I don't want to use 2 partitions for the squid because of backup/restore of all files. Best regards, Thomas Krause ----------------------------------------------------------- Dipl.-Ing. Thomas Krause DEH multimedia soft gmbh Nacamar-PoP Leipzig/Halle/Jena Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA32213; Fri, 8 Aug 1997 02:30:36 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 02:30:36 -0700 (PDT) Message-ID: <19970808112923.26887@admin.is.co.za> Date: Fri, 8 Aug 1997 11:29:23 +0200 From: Oskar Pearson <oskar@is.co.za> To: Michael Shoyher <msh@corbina.net> Cc: squid-users@nlanr.net Subject: Re: FD_SETSIZE and FreeBSD 2.1.7 References: <199708080839.MAA07986@ns.corbina.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69e In-Reply-To: <199708080839.MAA07986@ns.corbina.ru>; from Michael Shoyher on Fri, Aug 08, 1997 at 12:39:44PM +0400 Reply-To: oskar@is.co.za Resent-Message-ID: <"CrdX1UDYMbE.0.ve.FSkwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/138 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Fri, Aug 08, 1997 at 12:39:44PM +0400, Michael Shoyher wrote: > Is there any way to use more than 256 filedescriptors > on FreeBSD 2.1.7? I have to add this to the user's guide sometime: ------------------------------- In the "Changing operating system settings to enhance squid" chapter theres missing some information on how to enhance FreeBSD. The point with FreeBSD is, that parameters on how many FD's and buffers etc. are all configured by the MAXUSERS kernel configuration directive. A good value should be something like number of simultaneous connections multiplied by 1.5 or 2.0, depending of course what else the computer is doing. I'm using 2.0 because I can afford some spare memory on this matter. For example, putting MAXUSERS to 60 will get something about 1920 FD's and 960 Max processes. That should be enough for all. :-). - -- Antti Rytsola DLC Data Link Connections raccoon@dlc.fi ------------------------------- Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA02569; Fri, 8 Aug 1997 02:31:37 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 02:31:37 -0700 (PDT) Message-Id: <m0wwmHR-000i2jC@indiago.de> Date: Fri, 8 Aug 97 11:28 GMT+0100 To: Thomas Krause <tom@deh.de> From: Jan Jungnickel <octane@indiago.de> Cc: <squid-users@nlanr.net> Subject: Re: cache_swap > 6 GB In-Reply-To: <199708080925.LAA20149@chappi.deh.de> References: <199708080925.LAA20149@chappi.deh.de> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Mailer: Becky! ver 1.21 Resent-Message-ID: <"28AmoEpGLTD.0.Xd4.gTkwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/139 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Fri, 8 Aug 1997 11:23:29 +0200 "Thomas Krause" <tom@deh.de> wrote: > > I want to use a 7.3 GB cache partition for squid 1.1.14 > If the cache_swap value is larger then 6000000 (eg. 6300000) > I get the following message: > > WARNING: cache_swap (-2138734592 kbytes) is less than cache_mem (67108864 > bytes). As you see, this is a negative value, caused by the 32-Bit limitation ( about 4 Gig). > This will cause serious problems with your cache!!! > Change your configuration file. > > I don't want to use 2 partitions for the squid because > of backup/restore of all files. As i know no other solution, you`d have to use 2 partitions. If i am wrong please informe me. MfG, Jan Jungnickel indiago online - Technische Leitung http://www.indiago.de Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA26759; Fri, 8 Aug 1997 02:33:14 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 02:33:14 -0700 (PDT) Message-Id: <199708080932.TAA00051@valdez.mel.aone.com.au> X-Mailer: exmh version 2.0delta 6/3/97 To: ed@renishaw.co.uk cc: squid-users@nlanr.net Subject: Re: Starting Squid On Bootup In-reply-to: Your message of "Thu, 07 Aug 1997 16:56:59 GMT." <33E9F04B.54F8@renishaw.com> From: Lincoln Dale <ltd@aone.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 08 Aug 1997 19:32:20 +1000 Sender: ltd@aone.com.au Resent-Message-ID: <"f3lcSa1AfmE.0.U9.BVkwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/140 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > I have Squid running on a hp-ux 10.10 machine, and it works fine > when started from the command line. > > If I try to start with the normal rc scripts it just stops after > 'testing dns' > > Has anyone else experienced similar problems? its most likely nothing to do with the DNS test at all. i came across this one first with HP-UX 9.01. things that you put in the rc files sometimes just exit (for no apparent reason). rc scripts on HPUX that are started with #!/bin/sh can exit before the programs running in there have dis-associated themselves from the shell. either start squid using an 'at' job into the future (echo /usr/local/squid/bin/RunCache | at now + 1 minute'), start RunCache via nohup, or put a 'sleep 30' at the end of the rc script, after squid has been started. INN suffers a similar fate. cheers, lincoln. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id EAA27827; Fri, 8 Aug 1997 04:38:00 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 04:38:00 -0700 (PDT) From: "John Lauro" <jlauro@umich.edu> To: "squid-users@nlanr.net" <squid-users@nlanr.net>, "Thomas Krause" <tom@deh.de> Date: Fri, 08 Aug 97 07:30:53 -0500 Reply-To: "John Lauro" <jlauro@umich.edu> Priority: Normal X-Mailer: PMMail 1.51 For OS/2 UNREGISTERED SHAREWARE MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Subject: Re: cache_swap > 6 GB Message-ID: <37FCD11790@flint.umich.edu> Resent-Message-ID: <"uW2NKuP5AS4.0.l01.8Imwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/141 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Fri, 8 Aug 1997 11:23:29 +0200, Thomas Krause wrote: > >I want to use a 7.3 GB cache partition for squid 1.1.14 >If the cache_swap value is larger then 6000000 (eg. 6300000) >I get the following message: > >WARNING: cache_swap (-2138734592 kbytes) is less than cache_mem (67108864 >bytes). > This will cause serious problems with your cache!!! > Change your configuration file. > >I don't want to use 2 partitions for the squid because >of backup/restore of all files. > The cache_swap is in *MEGA*bytes. Set it to 6000 for 6GB. You asked for 6.3 tera-bytes of cache... Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id EAA03832; Fri, 8 Aug 1997 04:46:36 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 04:46:36 -0700 (PDT) Message-Id: <3.0.1.32.19970808214514.00da6f50@globec.com.au> X-Sender: sjudge@globec.com.au X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Fri, 08 Aug 1997 21:45:14 +1000 To: "Thomas Krause" <tom@deh.de> From: Steve Judge <sjudge@globec.com.au> Subject: Re: cache_swap > 6 GB Cc: squid-users@nlanr.net In-Reply-To: <199708080925.LAA20149@chappi.deh.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"ury9irBYzeA.0.KJ.2Rmwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/142 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 11:23 AM 8/8/97 +0200, you wrote: Thomas, The cache_swap size i think you'll find is in megabytes, so if you have a 7.3 gig cache partition it'd be in your squid.conf cache_swap 7300 > >I want to use a 7.3 GB cache partition for squid 1.1.14 >If the cache_swap value is larger then 6000000 (eg. 6300000) >I get the following message: Steve Judge Globec Internet Services Pty Ltd http://www.globec.com.au/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA02216; Fri, 8 Aug 1997 05:04:07 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 05:04:07 -0700 (PDT) From: Michael Shoyher <msh@corbina.net> Message-Id: <199708081201.QAA20942@ns.corbina.ru> Subject: Re: FD_SETSIZE and FreeBSD 2.1.7 To: oskar@is.co.za Date: Fri, 8 Aug 1997 16:01:49 +0400 (MSD) Cc: squid-users@nlanr.net In-Reply-To: <19970808112923.26887@admin.is.co.za> from "Oskar Pearson" at "Aug 8, 97 11:29:23 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"Upy0klog0tC.0.k_.Shmwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/143 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > On Fri, Aug 08, 1997 at 12:39:44PM +0400, Michael Shoyher wrote: > > Is there any way to use more than 256 filedescriptors > > on FreeBSD 2.1.7? > > The point with FreeBSD is, that parameters on how many FD's and buffers > etc. are all configured by the MAXUSERS kernel configuration directive. > A good value should be something like number of simultaneous connections > multiplied by 1.5 or 2.0, depending of course what else the computer is doing. > I'm using 2.0 because I can afford some spare memory on this matter. As I know FreeBSD 2.1.7 has problems with more than 256 descriptors in select(). And configure limits number of filedescriptors to 256. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA02363; Fri, 8 Aug 1997 05:18:54 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 05:18:54 -0700 (PDT) Message-ID: <19970808141635.18437@admin.is.co.za> Date: Fri, 8 Aug 1997 14:16:35 +0200 From: Oskar Pearson <oskar@is.co.za> To: Michael Shoyher <msh@corbina.net> Cc: squid-users@nlanr.net Subject: Re: FD_SETSIZE and FreeBSD 2.1.7 References: <19970808112923.26887@admin.is.co.za> <199708081201.QAA20942@ns.corbina.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69e In-Reply-To: <199708081201.QAA20942@ns.corbina.ru>; from Michael Shoyher on Fri, Aug 08, 1997 at 04:01:49PM +0400 Reply-To: oskar@is.co.za Resent-Message-ID: <"xhhNVWv8p4.0.Vn2.8vmwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/144 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi > > > Is there any way to use more than 256 filedescriptors > > > on FreeBSD 2.1.7? > > > > The point with FreeBSD is, that parameters on how many FD's and buffers > > etc. are all configured by the MAXUSERS kernel configuration directive. > > A good value should be something like number of simultaneous connections > > multiplied by 1.5 or 2.0, depending of course what else the computer is doing. > > I'm using 2.0 because I can afford some spare memory on this matter. > As I know FreeBSD 2.1.7 has problems with more than 256 descriptors in > select(). And configure limits number of filedescriptors to 256. I have personally seen a FreeBSD box with over 2000 concurrent TCP sessions to the squid port. I didn't configure it though. Oskar Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id GAA01253; Fri, 8 Aug 1997 06:01:53 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 06:01:53 -0700 (PDT) From: Andre Albsmeier <Andre.Albsmeier@mchp.siemens.de> Message-Id: <199708081300.PAA04159@curry.mchp.siemens.de> Subject: Re: get "Operation already in progress" when trying to use squid with socks In-Reply-To: <199708072346.JAA12205@diji.jna.com.au> from John Warburton at "Aug 8, 97 09:50:15 am" To: squid-users@nlanr.net Date: Fri, 8 Aug 1997 15:00:16 +0200 (CEST) X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"9P7vd8_HdBD.0.ZW1.vXnwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/145 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > Hi, > > we use socks 5 with our squid with no problems. > > > > The socks5 package itself runs great but with squid I had no succes. > > > > I socksified squid using the usual procedure: including socks.h, > > > > defining SOCKS and linking with the socks5 library. > > you no longer have to go through that rigmarole any more. We use the > > "runsocks" shell script which sets up the correct socks environment, > > so you don't have to modify the source. Our RunCache script has been > > modified. Instead of > > squid -sY $conf >> $logdir/squid.out 2>&1 > > we have > > /usr/local/bin/runsocks squid -sY $conf >> $logdir/squid.out > > 2>&1 > > > > pretty simple eh? Thanks to all how have responded to me. I have received the cool hint to use ssl_proxy together with the netscrape cache and so I have no need to socksify squid anymore :-) Thanks again -Andre Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA02315; Fri, 8 Aug 1997 07:02:17 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 07:02:17 -0700 (PDT) Date: Fri, 8 Aug 97 09:58:48 EDT From: "R. Scott Bailey" <sbailey@dsddi.eds.com> Subject: ftpget 1.1.14 truncates on space in filename To: squid-users@nlanr.net X-PRIORITY: 3 (Normal) X-Mailer: Chameleon 5.0, TCP/IP for Windows, NetManage Inc. Message-ID: <Chameleon.871048917.sbailey@wimpy.dsddi.eds.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"CLkPSATNFC7.0.zC1.bQowp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/146 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net If anybody else has found and fixed this problem already, I'd appreciate a pointer on a patch before I struggle through it myself. :-) ftpget (both 1.1.10 and 1.1.14) does not process file entries in a directory listing correctly when the filename contains a space (which can easily occur on Windows NT servers...). The reported name (and the name in the generated URL) is truncated at the space. Cheers, Scott ---------- R. Scott Bailey sbailey@dsddi.eds.com system manager, postmaster sbailey@xcc.mc.xerox.com EDS / ESC - Detroit 313-556-6011 [8/346] Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA06207; Fri, 8 Aug 1997 07:04:27 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 07:04:27 -0700 (PDT) Message-Id: <3.0.2.32.19970808145244.00c52450@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Fri, 08 Aug 1997 14:52:44 +0100 To: Robert Andersson <robert@gar.no>, Duane Wessels <wessels@nlanr.net> From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Netscape 4 ? Cc: squid-users@nlanr.net In-Reply-To: <mail_out.33ea6e3d.robert@gmail.gar.no> References: <199708072021.NAA14922@nlanr.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"apcB7x2dpCA.0.931.ZTowp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/147 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 02:54 08/08/97 +0200, Robert Andersson wrote: >It seems we agree on the NOCACHE case. What I don't understand is why >you want NOCACHE+IMS to always regard the cached object as valid and >thus bypass the refresh_pattern staleness checking. At least to me it >doesn't feel right that NOCACHE+IMS is less inclined to forward the >request than just a plain IMS FWIW, I think Robert has a good point - I would prefer his behaviour. Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA12316; Fri, 8 Aug 1997 07:06:47 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 07:06:47 -0700 (PDT) Message-Id: <3.0.2.32.19970808145856.00c521ec@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Fri, 08 Aug 1997 14:58:56 +0100 To: Adam Rigby <a.rigby@qut.edu.au>, squid-users@nlanr.net From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Cachemgr.cgi configuration In-Reply-To: <3.0.32.19970808101511.00556480@pop.qut.edu.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"MFEW4G8ERIB.0.2P1.bVowp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/148 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 10:15 08/08/97 +1000, Adam Rigby wrote: >After starting cachemgr.cgi and selecting any option from the drop down >list the following message appears eg. [snip 8< ] >ERROR > >Access Denied > >Sorry, you are not currently allowed to request: >cache_object://proxy.is.here/info@ [snip 8< ] Read the FAQ. Make sure you have the web server with the cachemgr.cgi script in the ACL. e.g. acl manager proto cache_object acl mywebserver src 1.2.3.4/255.255.255.255 http_access allow manager mywebserver http_access deny manager Make sure you have set up your web server security so that random people can't just access your cachemgr.cgi. Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA10362; Fri, 8 Aug 1997 07:27:20 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 07:27:20 -0700 (PDT) Message-Id: <3.0.2.32.19970808151107.00c535fc@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Fri, 08 Aug 1997 15:11:07 +0100 To: squid-users@nlanr.net From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Most Stable Squid In-Reply-To: <Pine.3.89.9708072301.A11663-0100000@teal.csn.net> References: <19970807133824.52715@texas.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"hKkzZOvtb13.0.UN.noowp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/149 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 23:56 07/08/97 -0600, David Hughes wrote: >What is the most stable current version of Squid? I would also like an answer to this. The web page at squid.nlanr.net still says the stable version is 1.1.11, and the most recent is 1.1.14, which implies 1.1.14 is unstable. But people seem to be using it happily, so what's the story? Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA18795; Fri, 8 Aug 1997 08:21:33 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 08:21:33 -0700 (PDT) Date: Fri, 8 Aug 1997 09:40:05 -0500 (CDT) From: Matt Isleb <kermit@reptile.clc.cc.il.us> To: Strange Alex <strange@mix.unicon.ru> cc: squid-users@nlanr.net Subject: Re: Logs rotating... In-Reply-To: <199708080449.LAA03388@mix.unicon.ru> Message-ID: <Pine.LNX.3.91.970808093828.16824A-100000@reptile.clc.cc.il.us> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"HQ3fanTEKT6.0.912.aapwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/150 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Fri, 8 Aug 1997, Strange Alex wrote: > Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ? > My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid` > nothing happend with my log files, even when I do so more than 10 times ;( > And please, tell my the proper way, how to rotate all squid log files . Perhaps the PID in the file is stale or not there? I have a handy killall program. I do 'killall -USR1 squid' Or rather.. LINUX has a handy killall program.. dunno 'bout other *nixes. Matt Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA16811; Fri, 8 Aug 1997 08:23:31 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 08:23:31 -0700 (PDT) Date: Fri, 8 Aug 1997 09:43:40 -0500 (CDT) From: Matt Isleb <kermit@reptile.clc.cc.il.us> To: squid-users@nlanr.net Message-ID: <Pine.LNX.3.91.970808094131.16852A-100000@reptile.clc.cc.il.us> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"fHBiZEzpp_4.0.gK6.hdpwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net Subject: X-Mailing-List: <squid-users@nlanr.net> archive/latest/151 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Umm could someone tell me why i keep getting a LOT of blank emails that look like this in my /var/spool/mail/kermit dir: >From squid-users-request@nlanr.net Thu Aug 7 10:38:44 1997 Status: O X-Status: I get about 5/day. Matt Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA02562; Fri, 8 Aug 1997 09:17:00 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 09:17:00 -0700 (PDT) Date: Fri, 8 Aug 1997 12:15:54 -0400 (EDT) From: Michael Pelletier <mikep@comshare.com> To: Jonathan Larmour <JLarmour@origin-at.co.uk> cc: squid-users@nlanr.net Subject: Re: Most Stable Squid In-Reply-To: <3.0.2.32.19970808151107.00c535fc@mailhost.origin-at.co.uk> Message-ID: <Pine.BSI.3.95.970808120910.16667Q-100000@valhalla.comshare.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"zMXsMh1KwQ9.0.qi1.bPqwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/153 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Fri, 8 Aug 1997, Jonathan Larmour wrote: > At 23:56 07/08/97 -0600, David Hughes wrote: > >What is the most stable current version of Squid? > > I would also like an answer to this. The web page at squid.nlanr.net still > says the stable version is 1.1.11, and the most recent is 1.1.14, which > implies 1.1.14 is unstable. But people seem to be using it happily, so > what's the story? I think that what the web page is indicating is that 1.1.11 is the latest *proven* stable version. This does not necessarily imply that 1.1.14 is unstable, just that it hasn't had time to be proven stable. I've been using 1.1.14+retry here and it's been up for the last nine days without a hitch, so I personally wouldn't call it "unstable." (According to the registration database, there's five other sites besides me running the retry patch, and I think the Austrian one has been up longer, since I did a restart to back out a couple of changes I made beyond the patch that Technodat is running.) -Mike Pelletier. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA04177; Fri, 8 Aug 1997 09:17:03 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 09:17:03 -0700 (PDT) Message-Id: <3.0.2.32.19970808170931.00c5797c@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Fri, 08 Aug 1997 17:09:31 +0100 To: Matt Isleb <kermit@reptile.clc.cc.il.us>, squid-users@nlanr.net From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Blank e-mail In-Reply-To: <Pine.LNX.3.91.970808094131.16852A-100000@reptile.clc.cc.il .us> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"JmGgzJ5mzuB.0.tm.DPqwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/152 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 09:43 08/08/97 -0500, Matt Isleb wrote: >Umm could someone tell me why i keep getting a LOT of blank emails that >look like this in my /var/spool/mail/kermit dir: > >>From squid-users-request@nlanr.net Thu Aug 7 10:38:44 1997 >Status: O >X-Status: Everything's fine here, which would imply its something local to you. Does other e-mail work 100%? You may also want to check your sendmail logs to see if anything weird is happening there. You should get two copies of this e-mail, one direct to you, and one via squid-users. Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA19867; Fri, 8 Aug 1997 09:59:02 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 09:59:02 -0700 (PDT) Sender: eculp@mexcom.net Message-ID: <33EB4F92.6E0A4CDF@mexcom.net> Date: Fri, 08 Aug 1997 11:55:46 -0500 From: Edwin Culp <eculp@mexcom.net> Organization: Mexico Communicates, S.C. X-Mailer: Mozilla 3.01Gold (X11; I; Linux 2.0.14 i586) MIME-Version: 1.0 To: Matt Isleb <kermit@reptile.clc.cc.il.us> CC: Strange Alex <strange@mix.unicon.ru>, squid-users@nlanr.net Subject: Re: Logs rotating... References: <Pine.LNX.3.91.970808093828.16824A-100000@reptile.clc.cc.il.us> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"dMZ8ZsQzyOE.0.Op2.u_qwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/154 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Matt Isleb wrote: > > On Fri, 8 Aug 1997, Strange Alex wrote: > > > Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ? > > My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid` > > nothing happend with my log files, even when I do so more than 10 times ;( > > And please, tell my the proper way, how to rotate all squid log files . > > Perhaps the PID in the file is stale or not there? I have a handy killall > program. I do 'killall -USR1 squid' Or rather.. LINUX has a handy > killall program.. dunno 'bout other *nixes. > > Matt I had this problem sometime back after moving the logs to /var/log rather than /usr/local/squid/logs. Everything worked fine except logfile rotation. I just moved the log files back to /usr/local/squid/logs because of a space shortage en /var and presto the rotation problem was solved. I never saw an error from the kill process doing it manually or with cron. It just didn't happen. Ed Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id KAA07381; Fri, 8 Aug 1997 10:52:05 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 10:52:05 -0700 (PDT) Message-Id: <199708081651.MAA24129@web.rural-net.com> From: "John E. Kozitzki" <johnk@rural-net.com> To: "Edwin Culp" <eculp@mexcom.net>, <squid-users@nlanr.net> Subject: Re: Logs rotating... Date: Fri, 8 Aug 1997 13:51:07 -0400 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1160 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"XDoFFpb0VrE.0.rk._nrwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/155 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net We have squid within the /usr/local/squid structure and we still can't seem to get the rotate logfile function to work. Thanx John Kozitzki Information Systems Manager B&W Co-op, Incorporated (http://www.bwcoop.com) RURAL-NET of Central Michigan (http://www.rural-net.com) B&W Farm Center / Case IH (http://www.bwcaseih.com) Phone:(517)842-3104 ext 207 Fax:(517)842-3108 E-Mail:johnk@rural-net.com Webpage:(http://www.rural-net.com/~johnk) ---------- > From: Edwin Culp <eculp@mexcom.net> > To: Matt Isleb <kermit@reptile.clc.cc.il.us> > Cc: Strange Alex <strange@mix.unicon.ru>; squid-users@nlanr.net > Subject: Re: Logs rotating... > Date: Friday, August 08, 1997 12:55 PM > > Matt Isleb wrote: > > > > On Fri, 8 Aug 1997, Strange Alex wrote: > > > > > Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ? > > > My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid` > > > nothing happend with my log files, even when I do so more than 10 times ;( > > > And please, tell my the proper way, how to rotate all squid log files Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id LAA03494; Fri, 8 Aug 1997 11:42:24 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 11:42:24 -0700 (PDT) Date: Fri, 8 Aug 97 14:31:23 EDT From: "R. Scott Bailey" <sbailey@dsddi.eds.com> Subject: RE: ftpget 1.1.14 truncates on space in filename To: squid-users@nlanr.net Cc: "R. Scott Bailey" <sbailey@dsddi.eds.com> X-PRIORITY: 3 (Normal) X-Mailer: Chameleon 5.0, TCP/IP for Windows, NetManage Inc. Message-ID: <Chameleon.871065511.sbailey@wimpy.dsddi.eds.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Resent-Message-ID: <"YOaInqqEHW1.0.Kc5.8Vswp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/156 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net This is a lame patch, but it works for now. After it's applied, if ftpget sees a DOS-format file listing and an entry contains more than 4 tokens, it assumes the filename must contain whitespace and uses the exact (unparsed) string as the name. My command of C is too feeble to come up with an efficient and general way of determining where the filename really starts, so I wouldn't call this patch a robust solution. This patch is against ftpget.c 1.1.14. Cheers anyway, Scott --- ftpget.c.dist Fri Aug 8 14:12:01 1997 +++ ftpget.c Fri Aug 8 14:18:43 1997 @@ -1854,7 +1854,7 @@ break; } - /* try it as a DOS listing */ + /* try it as a DOS/Windows listing */ if (n_tokens > 3 && p->name == NULL && sscanf(tokens[0], "%[0-9]-%[0-9]-%[0-9]", sbuf, sbuf, sbuf) == 3 && /* 04-05-70 */ @@ -1868,7 +1868,11 @@ } sprintf(sbuf, "%s %s", tokens[0], tokens[1]); p->date = xstrdup(sbuf); - p->name = xstrdup(tokens[3]); + if (n_tokens == 4) { /* Name in a single token */ + p->name = xstrdup(tokens[3]); + } else { /* Name apparently contains whitespace (95 or NT) */ + p->name = xstrdup(buf + 39); /* Evil assumption */ + } } /* Try EPLF format; carson@lehman.com */ if (p->name == NULL && buf[0] == '+') { ---------- R. Scott Bailey sbailey@dsddi.eds.com system manager, postmaster sbailey@xcc.mc.xerox.com EDS / ESC - Detroit 313-556-6011 [8/346] Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id MAA13033; Fri, 8 Aug 1997 12:28:46 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 12:28:46 -0700 (PDT) From: Joao Carlos Mendes Luis <jonny@mailhost.coppe.ufrj.br> Message-Id: <199708081918.QAA20543@gaia.coppe.ufrj.br> Subject: Re: FD_SETSIZE and FreeBSD 2.1.7 In-Reply-To: <199708081201.QAA20942@ns.corbina.ru> from Michael Shoyher at "Aug 8, 97 04:01:49 pm" To: msh@corbina.net (Michael Shoyher) Date: Fri, 8 Aug 1997 16:18:44 -0300 (EST) Cc: oskar@is.co.za, squid-users@nlanr.net X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"m3KMhE25kh5.0.2Z1.G6twp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/157 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net #define quoting(Michael Shoyher) // As I know FreeBSD 2.1.7 has problems with more than 256 descriptors in // select(). And configure limits number of filedescriptors to 256. I've make some changes here when I was running 2.1.7, but I could not check if that worked because I changed to 2.2 just a little after. First, edit your kernel config file and add a line like this: options "FD_SETSIZE=5120" Or any other number of descriptors you feel you need. Them, edit the squid sources and remove the limitation on 256 descriptors if it detects FreeBSD 2.1. Also, in rc.local, add some sysctl's to get more kern.maxfiles and kern.maxfilesperproc. If this really works, please tell us. Jonny -- Joao Carlos Mendes Luis jonny@gta.ufrj.br +55 21 290-4698 jonny@coppe.ufrj.br Universidade Federal do Rio de Janeiro UFRJ/COPPE/CISI PGP fingerprint: 29 C0 50 B9 B6 3E 58 F2 83 5F E3 26 BF 0F EA 67 Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id NAA21693; Fri, 8 Aug 1997 13:15:22 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 13:15:22 -0700 (PDT) From: Michael.Maravillo@philonline.com.ph Date: Sat, 9 Aug 1997 04:16:54 +0800 (GMT+0800) X-Sender: mmj@fiesta.philonline.com.ph To: Edwin Culp <eculp@mexcom.net> cc: Matt Isleb <kermit@reptile.clc.cc.il.us>, Strange Alex <strange@mix.unicon.ru>, squid-users@nlanr.net Subject: Re: Logs rotating... In-Reply-To: <33EB4F92.6E0A4CDF@mexcom.net> Message-ID: <Pine.LNX.3.96.970809041405.4461K-100000@fiesta.philonline.com.ph> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"gIa73s0nCp8.0.-34.3utwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/158 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Fri, 8 Aug 1997, Edwin Culp wrote: >Matt Isleb wrote: >> >> On Fri, 8 Aug 1997, Strange Alex wrote: >> >> > Can somebody explain me, what SIGUSR1 do in squid 1.1.10 ? >> > My logfile_rotate set to 10, but when I do kill -USR1 `cat squid.pid` >> > nothing happend with my log files, even when I do so more than 10 times ;( >> > And please, tell my the proper way, how to rotate all squid log files . >> >> Perhaps the PID in the file is stale or not there? I have a handy killall >> program. I do 'killall -USR1 squid' Or rather.. LINUX has a handy >> killall program.. dunno 'bout other *nixes. >> >> Matt > >I had this problem sometime back after moving the logs to /var/log >rather >than /usr/local/squid/logs. Everything worked fine except logfile >rotation. >I just moved the log files back to /usr/local/squid/logs because of a >space shortage en /var and presto the rotation problem was solved. I >never >saw an error from the kill process doing it manually or with cron. It >just didn't happen. I guess this is a case of incorrect file/directory permissions. The logs directory should be owned by the user running squid. Mike __ Michael J. Maravillo Philippines Online System Administrator PGP KeyID: 470AED9D InfoDyne, Incorporated http://www.philonline.com.ph/~mmj/ 4A N&M Bldg, 1184 Chino Roces, Makati Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id QAA23190; Fri, 8 Aug 1997 16:33:16 -0700 (PDT) Resent-Date: Fri, 8 Aug 1997 16:33:16 -0700 (PDT) Date: Sat, 9 Aug 1997 10:48:19 +1000 (EST) From: "Malcolm B.J. Garbutt" <mgarbutt@vic.ozland.net.au> X-Sender: mgarbutt@mildura.mildura.net.au To: Squid Users <squid-users@nlanr.net> Subject: DNS Fails Message-ID: <Pine.LNX.3.91.970809104702.17340B-100000@mildura.mildura.net.au> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"J_xbNA269V6.0.Q72.Cnwwp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/159 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, We have been encountering regular occurences of squid going into DNS Fail mode lately, this has not happened for some time, is there something that someone knows about that causes this to happen. _________________________________________________________ Malcolm Garbutt Network Operations- OZLAND.NET MILDURA.NET MURRAY.NET Office Ph. 03 50 212 991 Office Fax 03 50 212 932 Emergency Ph. 018 596 150 .....Bringing the World to You...... _________________________________________________________ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id BAA07068; Sat, 9 Aug 1997 01:27:18 -0700 (PDT) Resent-Date: Sat, 9 Aug 1997 01:27:18 -0700 (PDT) Message-Id: <199708090818.UAA18873@crash.ihug.co.nz> Comments: Authenticated sender is <dlim@pop.ihug.co.nz> From: "Daryl Lim" <dlim@ihug.co.nz> To: squid-users@nlanr.net Date: Sat, 9 Aug 1997 20:18:36 +0000 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Multiple Partitions On The Same Drive ?? Priority: normal X-mailer: Pegasus Mail for Win32 (v2.52) Resent-Message-ID: <"Fnyjrwa6Sd2.0.7S3.DX2xp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/160 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, I was wondering if it would improve performance if I had a few partitions on the same drive . Therefore Squid would spread the data across the partitions right ? This is only a theory though . And increase data output ? Daryl, | email : dlim@ihug.co.nz (Use This First) -=>Daryl Lim<=- -| email : daryl@genesis.pakuranga.school.nz | WWW : http://homepages.ihug.co.nz/~dlim/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA08062; Sat, 9 Aug 1997 02:02:09 -0700 (PDT) Resent-Date: Sat, 9 Aug 1997 02:02:09 -0700 (PDT) Date: Sat, 9 Aug 1997 19:00:30 +1000 (EST) Message-Id: <199708090900.TAA17279@valdez.mel.aone.com.au> X-Sender: ltd@valdez.mel.aone.com.au X-Mailer: Windows Eudora Version 2.1.1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: "Daryl Lim" <dlim@ihug.co.nz> From: Lincoln Dale <ltd@aone.com.au> Subject: Re: Multiple Partitions On The Same Drive ?? Cc: squid-users@nlanr.net Resent-Message-ID: <"mh6pyONWzD7.0.zR4.H73xp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/161 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net >I was wondering if it would improve performance if I had a few >partitions on the same drive . Therefore Squid would spread the data >across the partitions right ? This is only a theory though . And >increase data output ? any half-decent operating system (read: unix) will already endeavour to spread files out across the disk (its only things like DOS' FAT filesystem chose a scenario of always-allocate-at-beginning). evidence of this is doing a fsck and see how (little) a unix filesystem is fragmented. no -- i don't think you'd gain any performance by partitioning. you would, however, gain big-time on performance if you could stripe a partition across multiple disks. write performance on a box here, with a RAID-0 array across 3 fast-wide scsi busses of 2GB disks is in excess of 10 megabytes/sec. and that is out of the box, without any attempt to do any tuning. cheers, lincoln. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id QAA09213; Sat, 9 Aug 1997 16:58:49 -0700 (PDT) Resent-Date: Sat, 9 Aug 1997 16:58:49 -0700 (PDT) Message-ID: <33ED0120.9547CD45@gundaker.com> Date: Sat, 09 Aug 1997 18:45:37 -0500 From: Stuart Stock <stuart@gundaker.com> Reply-To: stuart@gundaker.com Organization: Gundaker Realtors X-Mailer: Mozilla 4.0 [en] (Win95; U) MIME-Version: 1.0 To: squid-users@nlanr.net Subject: Multiple ports and 1.1.14 X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"GbegeXAWhU5.0.5R.TAGxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/162 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net I've been running 1.1.10 with the multiple ports patch with great success. I'd like to upgrade to 1.1.14. Does it have multiple port support built in or do I have to wait for a patch? I'm not subscribed to the list so please reply to the From_ address. -- Stuart Stock stuart@gundaker.com Systems/Security Administrator http://www.gundaker.com Gundaker Realtors "Blowing a buffer as we speak" Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id SAA00598; Sat, 9 Aug 1997 18:01:46 -0700 (PDT) Resent-Date: Sat, 9 Aug 1997 18:01:46 -0700 (PDT) Date: Sun, 10 Aug 1997 10:59:21 +1000 (EST) Message-Id: <199708100059.KAA16348@valdez.mel.aone.com.au> X-Sender: ltd@valdez.mel.aone.com.au X-Mailer: Windows Eudora Version 2.1.1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: stuart@gundaker.com From: Lincoln Dale <ltd@aone.com.au> Subject: Re: Multiple ports and 1.1.14 Cc: squid-users@nlanr.net Resent-Message-ID: <"JLjdjLJvz69.0.ly4.9AHxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/163 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > I've been running 1.1.10 with the multiple ports patch with great > success. I'd like to upgrade to 1.1.14. Does it have multiple port > support built in or do I have to wait for a patch? I'm not subscribed > to the list so please reply to the From_ address. nope, 1.1.14 doesn't have the multiple-ports-patch built-in. it will, however, be part of the up-coming 1.2.xx squid series. the same patch as for 1.1.10 should apply (relatively) cleanly to 1.1.14, or at least with minimal modification (it should be fairly trivial). cheers, lincoln. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id EAA25473; Sun, 10 Aug 1997 04:31:31 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 04:31:31 -0700 (PDT) Message-ID: <19970810132542.52939@admin.is.co.za> Date: Sun, 10 Aug 1997 13:25:42 +0200 From: Oskar Pearson <oskar@is.co.za> To: squid-users@nlanr.net Subject: New squid docs out Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69e Reply-To: oskar@is.co.za Resent-Message-ID: <"vSIryWI1cR2.0.nl2.ELQxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/164 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hello all! There is a new version of the squid-docs out, and this time it has not one, but two sections written by volunteers! (from 0 pages to 2 in one month, and I have had another offer too!) The following sections are new: Log analysis and stats (Everyone say thank you to Mark Visser for this!) Preloading pages Optimisation/Optimization (And to Julian Anderson for this!) Bottlenecks and performance issues Malloc problems and memory usage System software you can use to analyze squid I have been trying to figure out a mirror policy... Here it is: The origin page (http://cache.is.co.za/squid/) will now keep the date of the last modification in the title. You can see if you have the latest version by comparing against this page. The latest version is always going to be at 'http://cache.is.co.za/squid-docs.tar.gz'. I have had numerous request for a 'paper-friendly' version of the document. It makes sense, so I am going to do it, but I am not sure when, since it involves re-typing most of the document, basically. Oskar Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id LAA14685; Sun, 10 Aug 1997 11:54:19 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 11:54:19 -0700 (PDT) Date: Sun, 10 Aug 1997 15:51:03 -0300 Message-Id: <199708101851.PAA16454@dess.tallships.istar.ca> subject: possible squid bug return-view-to: hclsmith@tallships.istar.ca (Hume Smith) to: squid-users@nlanr.net x-mailer: unicorn 0.46 (tk 4.2, tcl 7.6, windows Windows 95 4.0 intel) content-type: text/plain; charset=us-ascii priority: normal mime-version: 1.0 from: hclsmith@tallships.istar.ca (Hume Smith) organization: North End TTA Fans and Reed Organ Repairers Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by nlanr.net id LAA17350 Resent-Message-ID: <"fon4Cp7wUn5.0.wq2.esWxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/165 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net (I am not on this list.) i just upgraded 1.1.11-1.1.14, and both versions have a problem with <URL:http://spam.abuse.net/spam/> fetching through squid gives The requested URL /spam//spam/:80/spam/ was not found on this server. my browser (Opera 2.12) can fetch it directly just fine. (i have had similar problems with a few other sites - the request fails with a wildly mixed-up URL path. i don't know if it's Squid or something about the sites, that they don't like proxies, or what...) -- Hume Smith <URL:mailto:hclsmith@tallships.istar.ca> <URL:http://dess.tallships.istar.ca/%7Ehclsmith/> To hell with the information age, I want a wisdom age. - Bill Maxim Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id OAA03059; Sun, 10 Aug 1997 14:35:44 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 14:35:44 -0700 (PDT) Message-ID: <19970810233402.01509@admin.is.co.za> Date: Sun, 10 Aug 1997 23:34:02 +0200 From: Oskar Pearson <oskar@is.co.za> To: Hume Smith <hclsmith@tallships.istar.ca> Cc: squid-users@nlanr.net Subject: Re: possible squid bug References: <199708101851.PAA16454@dess.tallships.istar.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69e In-Reply-To: <199708101851.PAA16454@dess.tallships.istar.ca>; from Hume Smith on Sun, Aug 10, 1997 at 03:51:03PM -0300 Reply-To: oskar@is.co.za Resent-Message-ID: <"BdaTgmCvwp6.0.Rc7.UFZxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/166 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi > i just upgraded 1.1.11-1.1.14, and both versions have a problem with <URL:http://spam.abuse.net/spam/> fetching through squid gives > The requested URL /spam//spam/:80/spam/ was not found on this server. > my browser (Opera 2.12) can fetch it directly just fine. My guess is that it's almost certainly the browser getting confused. I have never seen this message and squid works perfectly with the above URL for me. I am running 1.NOVM.10. 1.1.11 has been out too long for this bug to have not shown up before now. Oskar Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id OAA22318; Sun, 10 Aug 1997 14:43:35 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 14:43:35 -0700 (PDT) Message-ID: <33ECC69F.178D@fua.br> Date: Sat, 09 Aug 1997 16:35:59 -0300 From: Ricardo Rios <rios@fua.br> X-Mailer: Mozilla 3.01Gold (Win95; I) MIME-Version: 1.0 To: squid-users@nlanr.net Subject: Squid cache_effective_user! Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"Tajk0OrmO16.0.cV3._NZxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/167 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, I'm having problems with the "cache_effective_user" option in squid version 1.NOVM.14 and 1.1.14. I tried to set this options to user "nobody" and group "nobody", but it did not work. I tried to set to user "squid" and group "adm", but again it did not work. Does somebody know what I can do to do squid work? Best regards, -- Ricardo Rios (rios@fua.br) Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id OAA17020; Sun, 10 Aug 1997 14:51:10 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 14:51:10 -0700 (PDT) Newsgroups: comp.infosystems.www.servers.unix Date: Sun, 10 Aug 1997 13:44:25 -0400 (EDT) From: Robert Vincent <inocent@exis.net> To: pam-list@redhat.com, morgan@parc.power.net, squid-users@nlanr.net, squid@nlanr.net, ingo@blank.pages.de Subject: Linux-Pam patch for Squid Message-ID: <Pine.LNX.3.95.970810133629.32178A-100000@sailfish.exis.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"m_vx0lDQXRF.0.214.4VZxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/168 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Wrote the following to use Linux-Pam with Squid. It's based in part on the mod_auth_pam writen by <ingo@blank.pages.de>. Feel free to copy, modify, and/or distribute. Next incremental improvement would be to make the service configurable via squid.conf. -------cut here------- diff -ur squid-1.NOVM.14/src/Makefile squid-1.NOVM.14.PAM/src/Makefile --- squid-1.NOVM.14/src/Makefile Sun Aug 10 06:42:08 1997 +++ squid-1.NOVM.14.PAM/src/Makefile Sun Aug 10 10:53:09 1997 @@ -7,7 +7,7 @@ # Uncomment and customize the following to suit your needs: # HOST_OPT = # -DCACHEMGR_HOSTNAME="getfullhostname()" -AUTH_OPT = # -DUSE_PROXY_AUTH=1 +AUTH_OPT = -DUSE_PROXY_AUTH=1 -DUSE_PROXY_AUTH_PAM=1 LOG_HDRS_OPT = # -DLOG_FULL_HEADERS=1 ICMP_OPT = # -DUSE_ICMP=1 DELAY_HACK = # -DDELAY_HACK=1 @@ -52,7 +52,7 @@ RANLIB = ranlib LN_S = ln -s PERL = /usr/local/bin/perl -CRYPTLIB = -lcrypt +CRYPTLIB = -lcrypt -lpam -ldl REGEXLIB = AC_CFLAGS = -O2 -fomit-frame-pointer -Wall LDFLAGS = -g diff -ur squid-1.NOVM.14/src/client_side.c squid-1.NOVM.14.PAM/src/client_side.c --- squid-1.NOVM.14/src/client_side.c Fri Jul 11 17:51:14 1997 +++ squid-1.NOVM.14.PAM/src/client_side.c Sun Aug 10 16:02:33 1997 @@ -295,6 +297,7 @@ icpProcessRequest(fd, icpState); } +#if !USE_PROXY_AUTH_PAM #if USE_PROXY_AUTH /* Check the modification time on the file that holds the proxy * passwords every 'n' seconds, and if it has changed, reload it @@ -429,6 +432,125 @@ return (sent_user); } #endif /* USE_PROXY_AUTH */ + +#else /*#if USE_PROXY_AUTH_PAM*/ + +#include <security/pam_appl.h> + +typedef struct { + char *name, *pw; +} auth_pam_userinfo; + +#define FAIL_DELAY 3000 + +int auth_pam_talker(int num_msg, + const struct pam_message **msg, + struct pam_response **resp, + void *appdata_ptr) +{ + unsigned short i = 0; + auth_pam_userinfo *userinfo = (auth_pam_userinfo*)appdata_ptr; + struct pam_response *response = 0; + + /* parameter sanity checking */ + if(!resp || !msg || !userinfo){ + debug(33,4,"auth_pam_talker: null parameters\n"); + return PAM_SUCCESS; + }; + + /* allocate memory to store response */ + response = xmalloc(num_msg * sizeof(struct pam_response)); + if(!response){ + debug(33,4,"auth_pam_talker: couldn't allocate memory for response\n"); + return PAM_SUCCESS; + }; + + /* copy values */ + for(i = 0; i < num_msg; i++) { + /* initialize to safe values */ + response[i].resp_retcode = 0; + response[i].resp = 0; + + /* select response based on requested output style */ + switch(msg[i]->msg_style) { + case PAM_PROMPT_ECHO_OFF: + debug(33,5,"auth_pam_talker: password is %s\n",userinfo->pw); + response[i].resp = xstrdup(userinfo->pw); + break; + case PAM_PROMPT_ECHO_ON: + debug(33,5,"auth_pam_talker: name is %s\n",userinfo->name); + response[i].resp = xstrdup(userinfo->name); + break; + default: + debug(33,4,"auth_pam_talker: unexpected conversation (%s)",msg[i]->msg); + return PAM_CONV_ERR; + break; + } + } + /* everything okay, set PAM response values */ + *resp = response; + return PAM_SUCCESS; +} + +const char * +proxyAuthenticate(const char *headers) +{ + char *pam_servicename="login"; + auth_pam_userinfo userinfo = { NULL, NULL }; + struct pam_conv conv_info = { &auth_pam_talker, &userinfo }; + pam_handle_t *pamh = NULL; + + char *s = NULL; + char *sent_userandpw = NULL; + char *passwd = NULL; + char *clear_userandpw = NULL; + int res; + LOCAL_ARRAY(char, sent_user, ICP_IDENT_SZ); + + /* Look for Proxy-authorization: Basic in the + * headers sent by the client + */ + if ((s = mime_get_header(headers, "Proxy-authorization:")) == NULL) { + debug(33, 4, "proxyAuthenticate: Can't find authorization header\n"); + return (dash_str); + } + /* Skip the 'Basic' part */ + s += strlen(" Basic"); + sent_userandpw = xstrdup(s); + strtok(sent_userandpw, "\n"); /* Trim trailing \n before decoding */ + clear_userandpw = uudecode(sent_userandpw); + xfree(sent_userandpw); + xstrncpy(sent_user, clear_userandpw, ICP_IDENT_SZ); + strtok(sent_user, ":"); /* Remove :password */ + userinfo.name = sent_user; + passwd = strstr(clear_userandpw, ":"); + passwd++; + userinfo.pw = passwd; + + if ((res=pam_start(pam_servicename,0,&conv_info,&pamh))!=PAM_SUCCESS){ + /* deny access;*/ + debug(33, 1, "proxyAuthenticate: failed to load service %s\n", pam_servicename); + pam_end(pamh, res); + xfree(clear_userandpw); + return (dash_str); + } + + pam_fail_delay(pamh, FAIL_DELAY); + + res = pam_authenticate(pamh, PAM_DISALLOW_NULL_AUTHTOK); + pam_end(pamh, res); + if(res!=PAM_SUCCESS){ + /* deny access;*/ + debug(33, 4, "proxyAuthenticate: authentication failed for user %s\n", sent_user); + xfree(clear_userandpw); + return (dash_str); + }; + + debug(33, 5, "proxyAuthenticate: user %s validated\n", sent_user); + xfree(clear_userandpw); + return (sent_user); +} +#endif /* USE_PROXY_AUTH_PAM */ void icpProcessExpired(int fd, void *data) Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id RAA21535; Sun, 10 Aug 1997 17:46:11 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 17:46:11 -0700 (PDT) Date: Mon, 11 Aug 1997 09:42:51 +1000 From: Adam Rigby <a.rigby@qut.edu.au> Subject: Re: Cachemgr.cgi configuration X-Sender: rigbya@pop.qut.edu.au To: squid-users@nlanr.net Cc: David J N Begley <d.begley@nepean.uws.edu.au>, Jonathan Larmour <JLarmour@origin-at.co.uk>, Ben Couldrey <ben@webnet.com.au> Message-id: <3.0.32.19970811094244.00599ea0@pop.qut.edu.au> MIME-version: 1.0 X-Mailer: Windows Eudora Pro Version 3.0 (32) Content-type: text/plain; charset="us-ascii" Resent-Message-ID: <"Jl8S_PKGpl4.0.YR5.h0cxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/169 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net You woukldn't beleive it, if you add a http_access line as follows "http_access allow localhost" and remove "!localhost" from the http_access deny line. Then everything is hunky dory, Fantastic, it bloody works. > acl localhost src 127.0.0.1/255.255.255.255 > acl all src 0.0.0.0/0.0.0.0 > > http_access allow manager localhost > http_access deny manager Thanks for the help everyone. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Adam Rigby Ph: 3864 1702 Network Programmer Fax: 3864 5272 Internetworking Software Services Computing Services e-mail:a.rigby@qut.edu.au Queensland University of Technology ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id SAA08259; Sun, 10 Aug 1997 18:00:24 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 18:00:24 -0700 (PDT) Message-ID: <33EE6328.3191CD55@schnet.edu.au> Date: Mon, 11 Aug 1997 10:56:08 +1000 From: Nick Clements <nickle@schnet.edu.au> Organization: Schoolsnet Aust.Pty.Ltd. X-Mailer: Mozilla 4.01 [en] (WinNT; I) MIME-Version: 1.0 To: "squid-users@nlanr.net" <squid-users@nlanr.net> Subject: (no subject) X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"9Mr9EoV_4n3.0.3Z4.PGcxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/170 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net unsubscribe nickle@schnet.edu.au Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id SAA18892; Sun, 10 Aug 1997 18:06:02 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 18:06:02 -0700 (PDT) Date: Sun, 10 Aug 1997 21:05:31 -0400 (EDT) From: Keith Heinemann <khman@neptune.new-era.net> To: squid-bugs@nlanr.net cc: squid-users@nlanr.net Subject: ftpget problem Message-ID: <Pine.LNX.3.95.970810210045.26125A-100000@neptune.new-era.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"vq797fwFN8D.0.BI2.eLcxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/171 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net (using squid 1.1.14) I've found that when using squid to access FTP sites, it doesn't properly convert the FTP 'readme' file to valid HTML. For example, ftp://tsx-11.mit.edu/ has some 'to get <file>.gz' entries. ftpget does not escape the < and > symbols as per HTML spec, thus my browser (netscape 3) tries to interpret them. I don't know if there are also other symbols that aren't being escaped correctly. Also, (squid-users), is it possible to disaply ftpget from converting ftp listings to HTMLscript? I believe the old harvest cached could, but I can't seem to find how to do this under squid. Thanks! -keith Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id SAA23863; Sun, 10 Aug 1997 18:25:58 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 18:25:58 -0700 (PDT) Date: Sun, 10 Aug 1997 21:18:58 -0400 (EDT) From: pstewart <pstewart@oncomdis.on.ca> To: squid-users@nlanr.net Subject: Connection Reset By Peer Problems Message-ID: <Pine.LNX.3.93.970810210449.18486B-100000@oncomdis.on.ca> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"tiJcg79Pcp1.0.oz3.Lecxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/172 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi there... Is there a retry or timeout value that can be set in Squid 1.NOVM.13 to stop it from doing the following: Connection Reset By Peer But if the user clicks on the URL link on the page, it loads fine... Anyone help? :) Paul -- Out the modem, through the SPARC, down the T3, off the router, past the frame-relay... nothing but Net. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id TAA21297; Sun, 10 Aug 1997 19:48:19 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 19:48:19 -0700 (PDT) Date: Sun, 10 Aug 1997 23:46:41 -0300 Message-Id: <199708110246.XAA18576@dess.tallships.istar.ca> subject: Re: possible squid bug return-view-to: hclsmith@tallships.istar.ca (Hume Smith) to: squid-users@nlanr.net x-mailer: unicorn 0.46 (tk 4.2, tcl 7.6, windows Windows 95 4.0 intel) in-reply-to: <19970810233402.01509@admin.is.co.za> content-type: text/plain; charset=us-ascii references: <19970810233402.01509@admin.is.co.za> <199708101851.PAA16454@dess.tallships.istar.ca> priority: normal mime-version: 1.0 from: hclsmith@tallships.istar.ca (Hume Smith) organization: North End TTA Fans and Reed Organ Repairers Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by nlanr.net id TAA24211 Resent-Message-ID: <"kvkhQfTmPn9.0.Ng3.tqdxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/173 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > My guess is that it's almost certainly the browser getting confused. i think you may be right. MesSIE can fetch it through the proxy fine; and apparently there's a cookie goes with it (abuse.net abusing cookies? wonderful :). perhaps it's the combination of cookie and proxy that messes Opera up. now how am i going to show them *this* bug... if i could figure out how to get Lynx to use a proxy i'd test it too :) anyway, that was my one qualm keeping me from broadening its use by Tallships and now i think i'll let more of us in on it... thanks! -- Hume Smith <URL:mailto:hclsmith@tallships.istar.ca> <URL:http://dess.tallships.istar.ca/%7Ehclsmith/> To hell with the information age, I want a wisdom age. - Bill Maxim Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id UAA16966; Sun, 10 Aug 1997 20:09:05 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 20:09:05 -0700 (PDT) Date: Mon, 11 Aug 1997 00:08:23 -0300 Message-Id: <199708110308.AAA18722@dess.tallships.istar.ca> subject: re: possible squid bug return-view-to: hclsmith@tallships.istar.ca (Hume Smith) to: squid-users@nlanr.net x-mailer: unicorn 0.46 (tk 4.2, tcl 7.6, windows Windows 95 4.0 intel) content-type: text/plain; charset=us-ascii priority: normal mime-version: 1.0 from: hclsmith@tallships.istar.ca (Hume Smith) organization: North End TTA Fans and Reed Organ Repairers Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by nlanr.net id UAA24328 Resent-Message-ID: <"4JuqqFmGCZF.0.Pf2.59exp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/174 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net WRT my trouble with <URL:http://spam.abuse.net/spam/> it works from lynx too. definitely an Opera problem. sorry to bother you :) -- Hume Smith <URL:mailto:hclsmith@tallships.istar.ca> <URL:http://dess.tallships.istar.ca/%7Ehclsmith/> To hell with the information age, I want a wisdom age. - Bill Maxim Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id UAA00391; Sun, 10 Aug 1997 20:56:07 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 20:56:07 -0700 (PDT) Date: Sun, 10 Aug 1997 23:44:49 -0400 From: "Armistead, Jason" <ARMISTEJ@oeca.otis.com> Subject: RE: possible squid bug To: "hclsmith@tallships.istar.ca" <hclsmith@tallships.istar.ca> Cc: "'squid-users@nlanr.net'" <squid-users@nlanr.net> Message-id: <c=US%a=_%p=UTC%l=UTC/OTUSIMC/000CCC41@otussimc.otis.com> X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.995.52 Encoding: 47 TEXT Resent-Message-ID: <"5_ZkjaDYLvD.0.H81.Tpexp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/175 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hume Lynx uses environment variables to configure Proxies. I'm not sure that it also handles Cookies too. Fote Macrides maintains a number of up-to date bug fixes and enhancements to Lynx (aside from being a Lynx developer) and his Lynx-version-FM code is always quite good. Check out Yahoo and do a search on Lynx. You'll find links to the current V2.7.1 version there. Anyhow, under VMS, the environment variable (logical name) is HTTP_PROXY or FTP_PROXY. Under Un*x I'd imagine it would be a case of setenv HTTP_PROXY "http://your.proxy:port/" and similar for FTP. These *may* need to be lower case - since I use a case insensitive O/S I can't tell from here. Good luck, and don't forget to check the Lynx FAQ Jason Armistead OEC-A, Minto Australia armistej@oeca.otis.com >---------- >From: hclsmith@tallships.istar.ca[SMTP:hclsmith@tallships.istar.ca] >Sent: Monday, 11 August 1997 12:46 >To: squid-users@nlanr.net >Subject: Re: possible squid bug > >> My guess is that it's almost certainly the browser getting confused. > >i think you may be right. MesSIE can fetch it through the proxy fine; and >apparently there's a cookie goes with it (abuse.net abusing cookies? >wonderful :). perhaps it's the combination of cookie and proxy that messes >Opera up. now how am i going to show them *this* bug... > >if i could figure out how to get Lynx to use a proxy i'd test it too :) > >anyway, that was my one qualm keeping me from broadening its use by Tallships >and now i think i'll let more of us in on it... thanks! > >-- >Hume Smith <URL:mailto:hclsmith@tallships.istar.ca> ><URL:http://dess.tallships.istar.ca/%7Ehclsmith/> > >To hell with the information age, I want a wisdom age. - Bill Maxim > > > Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id WAA21106; Sun, 10 Aug 1997 22:44:13 -0700 (PDT) Resent-Date: Sun, 10 Aug 1997 22:44:13 -0700 (PDT) Message-Id: <m0wxnE5-000cw5C@lx01.tietosavo.fi> Comments: Authenticated sender is <jt@pop3> From: "Jussi Torhonen" <jussi.torhonen@tietosavo.fi> Organization: Tietosavo Oy, Kuopio, Finland To: "Armistead, Jason" <ARMISTEJ@oeca.otis.com> Date: Mon, 11 Aug 1997 08:41:40 +0200 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: RE: possible squid bug CC: "'squid-users@nlanr.net'" <squid-users@nlanr.net> Priority: normal In-reply-to: <c=US%a=_%p=UTC%l=UTC/OTUSIMC/000CCC41@otussimc.otis.com> X-mailer: Pegasus Mail for Win32 (v2.54) Resent-Message-ID: <"r027XSZzR07.0.ow6.AOgxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/176 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On 10 Aug 97 at 23:44, squid-users@nlanr.net wrote: > Lynx uses environment variables to configure Proxies. I'm not sure that Alternative way is to setup those proxy settings in lynx.cfg config file. Jussi -*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Jussi Torhonen / Internet: jussi.torhonen@tietosavo.fi Tietosavo Oy / X.400: C=fi, ADMD=elisa, PRMD=tsnet, P.O.Box 1582 / O=tietosavo, S=torhonen, G=jussi FIN-70461 KUOPIO / AX.25: OH7DC@OH7RBA.#KUO.FIN.EU FINLAND / Tel: +358-17-193 231, Fax: +358-17-193 355 -*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id AAA21454; Mon, 11 Aug 1997 00:03:09 -0700 (PDT) Resent-Date: Mon, 11 Aug 1997 00:03:09 -0700 (PDT) X-Lotus-Fromdomain: NORTHERN TRUST From: "Leonard Hardy"<Leonard_Hardy@notes.ntrs.com> To: squid-users@nlanr.net Message-Id: <862564F0.00210479.00@chi-g01.ntrs.com> Date: Mon, 11 Aug 1997 01:00:39 -0500 Subject: Leonard Hardy is out of the office. Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Resent-Message-ID: <"xQI_B_qlfV5.0.PR4.4Zhxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/177 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net I will be out of the office from 08/11/97 until 08/15/97. I will be out of the bank on Friday, May 30. Please contact the following people with questions or problems: MoneyNet Tom Gardner x6288 Larry Wells x6888 NETS Jim Coats x7274 Todd Burke x6257 CLR/CASHLINE Plus Tim Krumholz x6094 ACCESS/CASHLINE Plus Jose Herrera x6197 George Callobre x2885 ATM/BASE24 Brian Westphal x1340 Tim Krumholz x6094 CIT Jeff Blaine x6838 Russ Younglove x5129 INTRANET Stephanie Cahn x2191 Jeff Blaine x6838 Anna Lamb x6171 Thanks, Len Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id DAA13660; Mon, 11 Aug 1997 03:17:25 -0700 (PDT) Resent-Date: Mon, 11 Aug 1997 03:17:25 -0700 (PDT) Message-ID: <19970811101334.21238.qmail@io.salford.ac.uk> From: mark@plato.salford.ac.uk (Mark Powell) Subject: Re: FD_SETSIZE and FreeBSD 2.1.7 To: squid-users@nlanr.net Date: 11 Aug 1997 11:13:30 +0100 X-Gated-To-News-By: NewsMaster <news@salford.ac.uk> Xref: io.salford.ac.uk salford.mailing-lists.squid-users:2634 Resent-Message-ID: <"h_Aw17zIbu6.0.uQ7.-Nkxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/178 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net In article <5sfs71$kh0$1@io.salford.ac.uk>, Joao Carlos Mendes Luis <jonny@mailhost.coppe.ufrj.br> wrote: >#define quoting(Michael Shoyher) >Also, in rc.local, add some sysctl's to get more kern.maxfiles and >kern.maxfilesperproc. I'm running 2.2-STABLE (or 2.2.2) and use ( sysctl -w kern.maxfiles=4096 sysctl -w kern.maxfilesperproc=4096 ) > /dev/null in one of the start-up scripts to set the max files in the kernel. >If this really works, please tell us. This really does work :) -- Mark Powell - Unix Information Officer - Clifford Whitworth Building A.I.S., University of Salford, Salford, Manchester, UK. Tel: +44 161 295 5936 Fax: +44 161 295 5555 Email: mark@salford.ac.uk finger mark@ucsalf.ac.uk (for PGP key) <A HREF="http://www.ucsalf.ac.uk/~mark/">Home Page</A> Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id DAA28285; Mon, 11 Aug 1997 03:19:47 -0700 (PDT) Resent-Date: Mon, 11 Aug 1997 03:19:47 -0700 (PDT) Message-ID: <19970811101902.21376.qmail@io.salford.ac.uk> From: mark@plato.salford.ac.uk (Mark Powell) Subject: Re: Squid cache_effective_user! To: squid-users@nlanr.net Date: 11 Aug 1997 11:19:00 +0100 X-Gated-To-News-By: NewsMaster <news@salford.ac.uk> Xref: io.salford.ac.uk salford.mailing-lists.squid-users:2635 Resent-Message-ID: <"Rk7bPoLiAe5.0.rv6.jSkxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/179 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net In article <5slcqs$d1u$1@io.salford.ac.uk>, Ricardo Rios <rios@fua.br> wrote: >Hi, > > I'm having problems with the "cache_effective_user" option in squid >version 1.NOVM.14 and 1.1.14. I tried to set this options to user >"nobody" and group "nobody", but it did not work. I tried to set to user >"squid" and group "adm", but again it did not work. Does somebody know >what I can do to do squid work? I've been having the same problems. I recently upgraded our site-wide cache from 1.1.10 to 1.1.14 with no problems. However, I'm setting up another cache to test proxy_auth-entication and have installed 1.1.14 on there with a fresh squid.conf file. I configured cache_effective_user and it would still complain about that parameter not being set. I tried to get debugging info. from squid, but to no avail. I put my own debugging messages in there recompiled it, installed it, ran it and then it worked fine ?? I then realised I'd forgotten to define HAVE_PROXY_AUTH in the src/Makefile so I added that and recompiled and installed. Now it doesn't work and my debugging messages are showing that cache_effective_user is now empty after parseConfigFile has been performed ??? Something seems amiss here? -- Mark Powell - Unix Information Officer - Clifford Whitworth Building A.I.S., University of Salford, Salford, Manchester, UK. Tel: +44 161 295 5936 Fax: +44 161 295 5555 Email: mark@salford.ac.uk finger mark@ucsalf.ac.uk (for PGP key) <A HREF="http://www.ucsalf.ac.uk/~mark/">Home Page</A> Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id DAA20957; Mon, 11 Aug 1997 03:42:10 -0700 (PDT) Resent-Date: Mon, 11 Aug 1997 03:42:10 -0700 (PDT) From: "James R Grinter" <jrg@blodwen.demon.co.uk> Message-Id: <9708111139.ZM977@renard.watching.org> Date: Mon, 11 Aug 1997 11:39:34 +0100 In-Reply-To: <19970810233402.01509@admin.is.co.za> "Re: possible squid bug" (Aug 10, 23:34) References: <199708101851.PAA16454@dess.tallships.istar.ca> <19970810233402.01509@admin.is.co.za> X-Subliminal: H is for Hypertext X-Thought: You wouldn't understand X-Mailer: Z-Mail (3.2.0 26oct94 MediaMail) To: oskar@is.co.za, Hume Smith <hclsmith@tallships.istar.ca>, squid-users@nlanr.net Subject: Re: possible squid bug Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Resent-Message-ID: <"3u-spJbxSJE.0.Bf7.unkxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/180 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Sun 10 Aug, 1997, Oskar Pearson <oskar@is.co.za> wrote: > <hclsmith@tallships.istar.ca> Hume Smith writes: >> i just upgraded 1.1.11-1.1.14, and both versions have a problem with <URL:http://spam.abuse.net/spam/> fetching through squid gives >> The requested URL /spam//spam/:80/spam/ was not found on this server. >> my browser (Opera 2.12) can fetch it directly just fine. > >My guess is that it's almost certainly the browser getting confused. I have >never seen this message and squid works perfectly with the above URL for >me. I am running 1.NOVM.10. 1.1.11 has been out too long for this bug to >have not shown up before now. in fact, it looks like an interaction bug between browser and their server (which happens to also be using Squid 1.1.5 as an accelerator) I can certainly get it to come up with odd redirections like 'Location: http://spam.abuse.net:2784/spam//spam/', especially if you miss off the trailing '/' from the first URL. James. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA32325; Mon, 11 Aug 1997 09:27:27 -0700 (PDT) Resent-Date: Mon, 11 Aug 1997 09:27:27 -0700 (PDT) Date: Mon, 11 Aug 1997 13:14:51 -0300 Message-Id: <199708111614.NAA21725@dess.tallships.istar.ca> subject: Re: possible squid bug return-view-to: hclsmith@tallships.istar.ca (Hume Smith) to: squid-users@nlanr.net x-mailer: unicorn 0.46 (tk 4.2, tcl 7.6, windows Windows 95 4.0 intel) in-reply-to: <9708111139.ZM977@renard.watching.org> content-type: text/plain; charset=us-ascii references: <9708111139.ZM977@renard.watching.org> <199708101851.PAA16454@dess.tallships.istar.ca> <19970810233402.01509@admin.is.co.za> priority: normal mime-version: 1.0 from: hclsmith@tallships.istar.ca (Hume Smith) organization: North End TTA Fans and Reed Organ Repairers Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by nlanr.net id JAA02888 Resent-Message-ID: <"JvF4UH_2OOB.0.nf.Wgpxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/181 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > I can certainly get it to come up with odd redirections like 'Location: > http://spam.abuse.net:2784/spam//spam/', especially if you miss off the > trailing '/' from the first URL. i refound a page that had been doing odd things: <URL:http://www.furnation.com/artists.html> itself works fine, but trying to follow any artist's anchor causes a bad rewrite. the anchors require a redirect (no traling solidus); again there are cookies; *but* you can correct the URL manually and it will work (unlike the spam page). Again it affects Opera but not MesSIE or Lynx. -- Hume Smith <URL:mailto:hclsmith@tallships.istar.ca> <URL:http://dess.tallships.istar.ca/%7Ehclsmith/> To hell with the information age, I want a wisdom age. - Bill Maxim Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id MAA30746; Mon, 11 Aug 1997 12:20:06 -0700 (PDT) Resent-Date: Mon, 11 Aug 1997 12:20:06 -0700 (PDT) Date: Mon, 11 Aug 1997 15:24:18 -0500 (AST) From: Mario de Mello Bittencourt Neto <mneto@buriti.com.br> To: Squid Cache <squid-users@nlanr.net> Subject: RunCache Message-ID: <Pine.LNX.3.95.970811150452.13400A-100000@zeus.buriti.com.br> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"kmU1DETcs9C.0.x74.cLsxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/182 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi Folks, I have installed a couple of days ago a rpm of squid (1.1.11 I think) and installed without problems. I have searched the rpm but could not find the RunCache script that I used to run in a rc.local (when the system was based on a Slackware distribution and squid 1.0.22). Can I use such script with 1.1.x ? Where should I call it (rc3.d?) ? Thanks. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id MAA09168; Mon, 11 Aug 1997 12:21:14 -0700 (PDT) Resent-Date: Mon, 11 Aug 1997 12:21:14 -0700 (PDT) Date: Mon, 11 Aug 1997 15:20:39 -0400 (EDT) From: Michael Pelletier <mikep@comshare.com> To: squid-users@nlanr.net Subject: FD/ipcache/etc. entries high water mark useful? Message-ID: <Pine.BSI.3.95.970811151728.494V-100000@valhalla.comshare.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"3SPoe7_MYi6.0.2y.SOsxp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/183 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net I was wondering if other people think that having entries in the cache information page for the high-water-mark values of things like filedescriptors, ip cache entries, memory usage, and other such things, would be useful? I'm thinking that that would help a cache manager determine where such values should be set, without having to wait for the cache to crash when it overshoots the limits. -Mike Pelletier. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id BAA12870; Tue, 12 Aug 1997 01:51:28 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 01:51:28 -0700 (PDT) Date: Tue, 12 Aug 1997 10:44:56 +0200 (MET DST) From: Jan Torreele <Jan.Torreele@belnet.be> Subject: Caching violates copyright? To: Squid Users <squid-users@nlanr.net> Message-ID: <Pine.3.89.9708121013.A10496-0100000@mahler.belnet.be> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"b1tZADCWjL8.0.L32.jA2yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/184 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi all, I got the attached message (see below) in my mailbox. "dalet.belnet.be" is our network wide parent cache. Although I don't think the request can be enforced, I'd like your opinions about this and if possible pointers to 'official' documents commenting on this issue. Thanks, -- Jan Torreele E-mail: Jan.Torreele@belnet.be BELNET Service Support Team Tel.: +32 (0) 2 / 238 34 70 ---------- Forwarded message ---------- Date: Sun, 10 Aug 1997 09:37:24 +0100 From:webmaster@pictographics.com To: sst@belnet.be Subject: caching of files Hello, I noticed this morning that your server "dalet.belnet.be" was caching every files on my server. While I appreciate your interest in these files, you will note from the terms and conditions listed on every single homepage on my website. Therefore, while I have no problem with your users visiting one or more of the pages served from my server, I DO NOT GIVE YOU PERMISSION to store these files on your server. I INSIST THAT YOU STOP COPYING FILES FROM MY SERVER IMMEDIATELY. I ALSO INSIST THAT YOU DELETE ALL FILES CURRENTLY STORED ON YOUR SERVER IMMEDIATELY. FAILURE TO DO SO WILL RESULT IN LEGAL ACTION. Thank you. Sincerely, Luke Dalla Bona President --------------------------------------------- Pictographics, Ltd. Serving and Designing Web Sites 69 Lansdowne Ave Sault Ste. Marie, Ontario P6B 1K5 1.705.949.4871 (phone/fax) http://www.pictographics.com Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA09935; Tue, 12 Aug 1997 02:03:04 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 02:03:04 -0700 (PDT) Message-Id: <199708120902.TAA16515@grey.interact.net.au> Subject: Re: Caching violates copyright? In-Reply-To: <Pine.3.89.9708121013.A10496-0100000@mahler.belnet.be> from Jan Torreele at "Aug 12, 97 10:44:56 am" To: Jan.Torreele@belnet.be (Jan Torreele) Date: Tue, 12 Aug 1997 19:02:00 +1000 (EST) Cc: squid-users@nlanr.net Reply-To: tom@interact.net.au From: "Tom Minchin" <tom@interact.net.au> X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"7zVkBRAn-a8.0.y_4.yQ2yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/185 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Easily fixed. Just deny access to pictographics.com to your users. A more rational approach (in face of a frothing webmaster) is to use cache_stoplist on his domain. Then your users will only be proxying his pages, not caching them. Alternatively, since you're in Belgium and he's in the US is to ignore him. tom@interact.net.au > Hi all, > > I got the attached message (see below) in my mailbox. > "dalet.belnet.be" is our network wide parent cache. Although I > don't think the request can be enforced, I'd like your opinions about > this and if possible pointers to 'official' documents commenting on this > issue. > > Thanks, > > -- > Jan Torreele E-mail: Jan.Torreele@belnet.be > BELNET Service Support Team Tel.: +32 (0) 2 / 238 34 70 > > ---------- Forwarded message ---------- > Date: Sun, 10 Aug 1997 09:37:24 +0100 > From:webmaster@pictographics.com > To: sst@belnet.be > Subject: caching of files > > Hello, > > I noticed this morning that your server "dalet.belnet.be" was caching every > files on my server. While I appreciate your interest in these files, you > will note from the terms and conditions listed on every single homepage on > my website. > > Therefore, while I have no problem with your users visiting one or more of > the pages served from my server, I DO NOT GIVE YOU PERMISSION to store > these files on your server. > > I INSIST THAT YOU STOP COPYING FILES FROM MY SERVER IMMEDIATELY. I ALSO > INSIST THAT YOU DELETE ALL FILES CURRENTLY STORED ON YOUR SERVER > IMMEDIATELY. > > FAILURE TO DO SO WILL RESULT IN LEGAL ACTION. > > Thank you. > > Sincerely, > > Luke Dalla Bona > President > > > --------------------------------------------- > Pictographics, Ltd. Serving and Designing Web Sites > 69 Lansdowne Ave > Sault Ste. Marie, Ontario P6B 1K5 > 1.705.949.4871 (phone/fax) > http://www.pictographics.com > > Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA21345; Tue, 12 Aug 1997 02:12:07 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 02:12:07 -0700 (PDT) Message-ID: <33F028FE.B2B38F3E@wisper.net> Date: Tue, 12 Aug 1997 10:12:30 +0100 From: Leigh Porter <leigh@wisper.net> Reply-To: leigh@wisper.net Organization: Wisper Bandwidth Plc. X-Mailer: Mozilla 4.01 [en] (WinNT; I) MIME-Version: 1.0 To: Jan Torreele <Jan.Torreele@belnet.be> CC: Squid Users <squid-users@nlanr.net> Subject: Re: Caching violates copyright? X-Priority: 3 (Normal) References: <Pine.3.89.9708121013.A10496-0100000@mahler.belnet.be> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"Js-cy6ua9uC.0.uX4.9Z2yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/186 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net #include <stddisc.h> Jan Torreele wrote: > Hello, > > I noticed this morning that your server "dalet.belnet.be" was caching > every > files on my server. While I appreciate your interest in these files, > you > will note from the terms and conditions listed on every single > homepage on > my website. > > Therefore, while I have no problem with your users visiting one or > more of > the pages served from my server, I DO NOT GIVE YOU PERMISSION to store > > these files on your server. > > I INSIST THAT YOU STOP COPYING FILES FROM MY SERVER IMMEDIATELY. I > ALSO > INSIST THAT YOU DELETE ALL FILES CURRENTLY STORED ON YOUR SERVER > IMMEDIATELY. > > FAILURE TO DO SO WILL RESULT IN LEGAL ACTION. This is quite unbelieveable! Whenever anybody uses a modern browser then objectsfrom a WWW site will be held in a local cache anyway, whats he gunner do? Sue the world? Also, it is not you who are copying files from his server, it is your users - he could just as well say that he does not want any of his files to be stored in the memory of your router :) Aparrently, at a meeting in Amsterdam some Legal people said that since the files in question would be freely avaliable on the net and since the cache server is a public service designed to make that public information more avaliable then the question of copyright violations is absurd. (Maybe not the exact words, taken from a third party!) I think that the chap first needs to be politely told what a cache server is and what it does as i suspect they are not too sure of this, the fact that it actually makes their site faster and more pleasurable for the end users to use is probobly oblivious to them. In any event, these sort of things have all too often been blown out of all proportion by people trying to protect themselves (As I suspect the above sompany is doing) whilst not quite understanding that what they object to is actually helping them. Funny old world! -- Leigh Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id EAA19700; Tue, 12 Aug 1997 04:45:37 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 04:45:37 -0700 (PDT) Message-Id: <3.0.3.32.19970812074122.006f9254@monroe.lib.mi.us> X-Sender: kip@monroe.lib.mi.us X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Tue, 12 Aug 1997 07:41:22 -0400 To: squid-users@nlanr.net From: Kip DeGraaf <kip@monroe.lib.mi.us> Subject: Re: Caching violates copyright? In-Reply-To: <199708120902.TAA16515@grey.interact.net.au> References: <Pine.3.89.9708121013.A10496-0100000@mahler.belnet.be> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"8OI-WC6dmQB.0.3k5.pm4yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/187 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 07:02 PM 8/12/97 +1000, Tom Minchin wrote: >Easily fixed. Just deny access to pictographics.com to your users. A more >rational approach (in face of a frothing webmaster) is to use cache_stoplist >on his domain. Then your users will only be proxying his pages, not caching >them. Here's the relevant section from http://www.pictographics.com/termsandconditions: (Would someone explain to me where in this legalspeak that it implies caching proxies are not okay to use?) RESTRICTIONS ON USE Material from www.pictographics.com and any other World Wide Web site owned, operated, licensed, or controlled by Pictographics Ltd. or any of its related, affiliated, or subsidiary companies, may not be copied or distributed, or republished, uploaded, posted, or transmitted in any way, without the prior written consent of Pictographics Ltd. EXCEPT in the following circumstance: You may download one copy of the material on one computer for your personal, non-commercial home or educational use only, provided you do not delete or change any copyright, trademark, or other proprietary notices. Modification or use of the materials for any other purpose violates Pictographics Ltd. intellectual property rights. The material in this site is provided for lawful informational purposes only. Prior written consent must be obtained for any commercial use of data provided on this website. Please forward any requests in writing to Pictographics Ltd. at 69 Lansdowne Avenue, Sault Ste. Marie, Ontario Canada P6B 1K5 Attn: Legal Dept. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA22186; Tue, 12 Aug 1997 05:06:21 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:06:21 -0700 (PDT) Message-Id: <199708121203.WAA24979@grey.interact.net.au> Subject: Re: Caching violates copyright? In-Reply-To: <3.0.3.32.19970812074122.006f9254@monroe.lib.mi.us> from Kip DeGraaf at "Aug 12, 97 07:41:22 am" To: kip@monroe.lib.mi.us (Kip DeGraaf) Date: Tue, 12 Aug 1997 22:03:58 +1000 (EST) Cc: squid-users@nlanr.net Reply-To: tom@interact.net.au From: "Tom Minchin" <tom@interact.net.au> X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"b6ssqbwAfk7.0.aG5.A55yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/188 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > At 07:02 PM 8/12/97 +1000, Tom Minchin wrote: > >Easily fixed. Just deny access to pictographics.com to your users. A more > >rational approach (in face of a frothing webmaster) is to use cache_stoplist > >on his domain. Then your users will only be proxying his pages, not caching > >them. > Here's the relevant section from > http://www.pictographics.com/termsandconditions: > (Would someone explain to me where in this legalspeak that it implies > caching proxies are not okay to use?) > > You may download one copy of the material on one computer for your > personal, non-commercial home or educational use only, provided you do not > delete or > change any copyright, trademark, or other proprietary notices. Modification > or use of the materials for any other purpose violates Pictographics Ltd. > intellectual property rights. The material in this site is provided for > lawful informational purposes only. Well, just looking at his legal crap it could appear that anyone who wanted to view his homepages (it's just a lame web 'mall' - about 20 clients) with the view to buy something from him/them isn't able to unless they apply to use it for commercial purposes. Reading it his way, he believes that the cached objects is 'one copy of the material' and your proxy server isn't 'personal, non-commercial home or educational use only'. Since the site isn't a landmark site on the internet, just put in a redirect to www.playboy.com or www.altavista.digital.com. Your users probably won't notice, and may even be grateful in the end. His beef, which he should come out right and say, is that his counters and page access stats are pitiful and the 10 or so cached hits you got are deeply offensive to him. tom@interact.net.au Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA26169; Tue, 12 Aug 1997 05:08:15 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:08:15 -0700 (PDT) Sender: jamie@kb.netchannel.co.uk Message-ID: <33F05F5E.7F02A164@zusammen.com> Date: Tue, 12 Aug 1997 13:04:30 +0000 From: Jamie Wilson <jamie@zusammen.com> Organization: Zusammen X-Mailer: Mozilla 3.01 (X11; I; Linux 2.0.30 i586) MIME-Version: 1.0 To: Jan Torreele <Jan.Torreele@belnet.be> CC: Squid Users <squid-users@nlanr.net> Subject: Re: Caching violates copyright? References: <Pine.3.89.9708121013.A10496-0100000@mahler.belnet.be> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"BZu1YZN8JTA.0.Tb3._75yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/189 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Jan Torreele wrote: > > Hi all, > > I got the attached message (see below) in my mailbox. > "dalet.belnet.be" is our network wide parent cache. Although I > don't think the request can be enforced, I'd like your opinions about > this and if possible pointers to 'official' documents commenting on this > issue. Block access to the site in question, wait for anything from his site to flush out of your cache, and let the person in question know that you have complied with his request. I can't see how this would make him happy, but, well, it takes all kinds, I guess. :-/ I had a look at the site in question and noticed that there were some Beatles samples on the site - I doubt that there would be too many sites in the world allowed to distribute Beatles stuff (or use the Beatles logo, for that matter). Anyone bored enough to contact The Beatles management? Ho hum. ...j Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA04427; Tue, 12 Aug 1997 05:18:58 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:18:58 -0700 (PDT) Date: Tue, 12 Aug 1997 14:17:44 +0200 (MET DST) From: Jan Torreele <Jan.Torreele@belnet.be> Subject: Re: Caching violates copyright? To: Tom Minchin <tom@interact.net.au> cc: squid-users@nlanr.net In-Reply-To: <199708120902.TAA16515@grey.interact.net.au> Message-ID: <Pine.3.89.9708121444.A14617-0100000@mahler.belnet.be> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"9vbLRf_1JXA.0.Qg5.DI5yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/190 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Tom Minchin wrote: > Easily fixed. Just deny access to pictographics.com to your users. This is not something we want to do. > A more > rational approach (in face of a frothing webmaster) is to use cache_stoplist > on his domain. Then your users will only be proxying his pages, not caching > them. Sure, but it defeats the whole purpose of having a cache in the first place. > > Alternatively, since you're in Belgium and he's in the US is to ignore him. :-) my thinking too. But the question really is, *is* there a copyright violation going on here? After all, he could also complain that there are illegal copies of is stuff sitting in the buffers of my routers, etc. In my opinion I'm just providing an infrastructure that allow my users to efficiently access documents publicly accessible on the Internet. -- Jan Torreele E-mail: Jan.Torreele@belnet.be BELNET Service Support Team Tel.: +32 (0) 2 / 238 34 70 Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA20129; Tue, 12 Aug 1997 05:23:44 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:23:44 -0700 (PDT) Date: Tue, 12 Aug 1997 22:20:25 +1000 (EST) From: David J N Begley <david@avarice.nepean.uws.edu.au> Reply-To: David J N Begley <d.begley@nepean.uws.edu.au> To: Kip DeGraaf <kip@monroe.lib.mi.us> cc: squid-users@nlanr.net Subject: Re: Caching violates copyright? In-Reply-To: <3.0.3.32.19970812074122.006f9254@monroe.lib.mi.us> Message-ID: <Pine.LNX.3.96.970812221333.13528B-100000@avarice.nepean.uws.edu.au> X-No-Archive: Yes MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"P1dZ8WYMAP5.0.Ri3.tM5yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/191 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Kip DeGraaf wrote: > (Would someone explain to me where in this legalspeak that it implies > caching proxies are not okay to use?) Bearing in mind that I'm no lawyer, I'd guess the relevant sections are: > Material from www.pictographics.com ... may not be copied or > distributed, or republished, uploaded, posted, or transmitted in any > way ... You may download one copy of the material on one computer for your > personal, non-commercial home or educational use only ... By storing the information in a proxy's cache, it could be argued that the proxy administrator is permitting, aiding or actually executing the further distribution or republication of said cached material, outside the terms granted by the copyright owner. (Washes out mouth.) Of course, there exist "ways and means" of making your material non-cacheable - perhaps this site should consider using them. Cheers.. dave Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA17383; Tue, 12 Aug 1997 05:34:14 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:34:14 -0700 (PDT) From: "Michael 'Moose' Dinn" <dinn@moose.webworks.ca> Message-Id: <199708121233.JAA01637@moose.webworks.ca> Subject: Re: Caching violates copyright? To: Jan.Torreele@belnet.be (Jan Torreele) Date: Tue, 12 Aug 1997 09:33:24 -0300 (ADT) Cc: squid-users@nlanr.net In-Reply-To: <Pine.3.89.9708121444.A14617-0100000@mahler.belnet.be> from "Jan Torreele" at Aug 12, 97 02:17:44 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"A4fWMKG4Sz9.0.sB4.nW5yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/192 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Jan Torreele wrote: > But the question really is, *is* there a copyright violation going > on here? After all, he could also complain that there are illegal copies > of is stuff sitting in the buffers of my routers, etc. In my opinion I'm > just providing an infrastructure that allow my users to efficiently > access documents publicly accessible on the Internet. > Why not mail the folks who are running the legality-laden web site and see what their policy is on caches? -- Michael 'Moose' Dinn | Webworks Worldwide Inc. | dinn@webworks.ca If you have a 1972 Mustang Convertible I want to hear from you! Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA04424; Tue, 12 Aug 1997 05:51:37 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:51:37 -0700 (PDT) Message-Id: <3.0.1.32.19970812134906.0113c8f0@hydra.cc.ruu.nl> X-Sender: henny@hydra.cc.ruu.nl X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Tue, 12 Aug 1997 13:49:06 +0100 To: Jan Torreele <Jan.Torreele@belnet.be>, Tom Minchin <tom@interact.net.au> From: Henny Bekker <Henny.Bekker@cc.ruu.nl> Subject: Re: Caching violates copyright? Cc: squid-users@nlanr.net In-Reply-To: <Pine.3.89.9708121444.A14617-0100000@mahler.belnet.be> References: <199708120902.TAA16515@grey.interact.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"h8EzauRi7N5.0.OI.2l5yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/193 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 02:17 PM 8/12/97 +0200, Jan Torreele wrote: [...Stuff deleted...] >But the question really is, *is* there a copyright violation going >on here? After all, he could also complain that there are illegal copies >of is stuff sitting in the buffers of my routers, etc. In my opinion I'm >just providing an infrastructure that allow my users to efficiently >access documents publicly accessible on the Internet. > In my opiniom a caching-server (in this case a Web caching-server but it also applies to other caching such as ordinary disk-caching servers) can be considered as a (very) long network cable.. Cheers, Henny NB This applies _not_ for mirror servers.. ----------------------------------------------------------------------------- E-Mail: Henny.Bekker@cc.ruu.nl ! Disclaimer: The main obstacle to progress http http://www.ruu.nl/~henny ! progress is not ignorance,but PTT: Voice: +31 30 2536971 Fax: +31 30 2531633 ! the illusion of knowledge X400: /G=Henny/S=Bekker/OU=cc/O=ruu/PRMD=surf/ADMD=400net/C=nl o Paper: H.J.Bekker, Utrecht University, Computer Centre _ /- _ Po Box 80011, 3508 TA Utrecht Nederland (_) > (_) ----------------------------------------------------------------------------- Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA14292; Tue, 12 Aug 1997 05:51:37 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:51:37 -0700 (PDT) Message-Id: <3.0.3.32.19970812084918.00cb5da0@monroe.lib.mi.us> X-Sender: kip@monroe.lib.mi.us X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Tue, 12 Aug 1997 08:49:18 -0400 To: squid-users@nlanr.net From: Kip DeGraaf <kip@monroe.lib.mi.us> Subject: Re: Caching violates copyright? In-Reply-To: <199708121233.JAA01637@moose.webworks.ca> References: <Pine.3.89.9708121444.A14617-0100000@mahler.belnet.be> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"_sAeDuwM2YB.0.ja3.8l5yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/194 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Already have, just waiting for a reply. At 09:33 AM 8/12/97 -0300, Michael 'Moose' Dinn wrote: >Why not mail the folks who are running the legality-laden web site and see >what their policy is on caches? Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA23488; Tue, 12 Aug 1997 05:51:39 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:51:39 -0700 (PDT) Message-ID: <XFMail.970812144945.finken@nentec.de> X-Mailer: XFMail 1.2-alpha [p0] on SunOS Sender: finken@nentec.de Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <3.0.2.32.19970807202005.00c58a94@mailhost.origin-at.co.uk> Date: Tue, 12 Aug 1997 14:49:45 +0200 (MET DST) Organization: Maniac Coffee Drinkers From: Michael Finken <finken@nentec.de> To: squid-users@nlanr.net Subject: Re: squid 1.1.11 not caching anything Resent-Message-ID: <"3r-auODWQm5.0.835.Kl5yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/195 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On 07-Aug-97 Jonathan Larmour wrote: >At 20:57 07/08/97 +0200, Michael Finken wrote: >>I'm running squid 1.1.11 on Linux. Since about one week squid doesn't cache >>anything. I can find only 'RELEASE' messages in store.log and cache/log is >>empty. I've already removed the cache and restartet squid, no help. >>As far as I can see the configuration file hasn't been changed. > >What about cache.log and access.log in the logs/ directory? Info from there >is more useful for this kind of problem generally. The problem has been solved. When running squid with high debug levels I saw that all objects were considered too big to be cached. I saw that maxobjectsize in squid.conf was way to high (81920000), the scanf() call interpreting the number returned nonsense, probably 0. It would be nice if the configuration code could perform some sanity checks and issue a warning if something looks fishy... Michael ------------------------------------------------------------------------------- Michael Finken NENTEC Netzwerktechnologie GmbH Tel.: +49 721 9495-0 76227 Karlsruhe/Germany - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - WINDOWS... It's from those who gave us EDLIN... Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA20584; Tue, 12 Aug 1997 05:53:21 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 05:53:21 -0700 (PDT) From: "John Lauro" <jlauro@umich.edu> To: "squid-users@nlanr.net" <squid-users@nlanr.net>, "Jan Torreele" <Jan.Torreele@belnet.be> Date: Tue, 12 Aug 97 08:47:27 -0500 Reply-To: "John Lauro" <jlauro@umich.edu> Priority: Normal X-Mailer: PMMail 1.51 For OS/2 UNREGISTERED SHAREWARE MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Subject: Re: Caching violates copyright? Message-ID: <99467042AD@flint.umich.edu> Resent-Message-ID: <"ueYFj1R5LqB.0.8V3.Ro5yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/196 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997 10:44:56 +0200 (MET DST), Jan Torreele wrote: >Hi all, > >I got the attached message (see below) in my mailbox. >"dalet.belnet.be" is our network wide parent cache. Although I >don't think the request can be enforced, I'd like your opinions about >this and if possible pointers to 'official' documents commenting on this >issue. > There were already several comments on this, so I'll try to only point out some new ones... >I noticed this morning that your server "dalet.belnet.be" was caching every >files on my server. While I appreciate your interest in these files, you >will note from the terms and conditions listed on every single homepage on >my website. Is anyone running anything to grab *every* file from his server? (Such as one of the prefetch methods that have been talked about before on this list)?) Perhaps a user, and you are just stuck in between holding the files? Also, I would write back that the request is unreasonable given that the internet has standard ways to set a file to not to be cacheable or to expire (which squid supports unless you patched it...) and that *his* failure to follow the standards make *him* liable for them being cached. Of course this does open up the issue for those sites that are intentionally setting the headers and have some sort of legal speak and those using a patched version of squid to cache the pages anyways... Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id GAA20530; Tue, 12 Aug 1997 06:21:56 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 06:21:56 -0700 (PDT) Date: Tue, 12 Aug 1997 07:42:58 -0500 (CDT) From: Matt Isleb <kermit@reptile.clc.cc.il.us> To: Tom Minchin <tom@interact.net.au> cc: Jan Torreele <Jan.Torreele@belnet.be>, squid-users@nlanr.net Subject: Re: Caching violates copyright? In-Reply-To: <199708120902.TAA16515@grey.interact.net.au> Message-ID: <Pine.LNX.3.91.970812074131.412A-100000@reptile.clc.cc.il.us> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"y_cx1Ryuft.0.0b3.ZD6yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/197 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Tom Minchin wrote: > Easily fixed. Just deny access to pictographics.com to your users. A more > rational approach (in face of a frothing webmaster) is to use cache_stoplist > on his domain. Then your users will only be proxying his pages, not caching > them. > > Alternatively, since you're in Belgium and he's in the US is to ignore him. Actually he is in Canada. Don't let him hear you call him American, Canadians can get sensative about that kind of thing :) Matt Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id GAA27514; Tue, 12 Aug 1997 06:22:48 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 06:22:48 -0700 (PDT) Date: Tue, 12 Aug 1997 21:21:33 +0800 (CST) From: Ronald Wiplinger <ronald@linkou.trace.com.tw> To: Tom Minchin <tom@interact.net.au> cc: Jan Torreele <Jan.Torreele@belnet.be>, squid-users@nlanr.net Subject: Re: Caching violates copyright? In-Reply-To: <199708120902.TAA16515@grey.interact.net.au> Message-ID: <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"315sljkAqJE.0.CR1.UE6yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/198 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net I think the site should do something, if he don't like to cache, he can expire the page, like M$ it do. He also can give a message on his site that proxy is to turn off. Finally he should read some books, that he can get the meaning of the Internet, bandwith usage, .... Maybe he will need somebody who can read the book for him. He should get back to his job he was doing still last month ;-) We have redirected his site to www.disney.com. and added him to the bounce list (sendmail). I don't deal with such newbies. bye Ronald Wang's Trace International (Taipei) http://www.trace.com.tw Tel: +886 2 609-0652, Handy: +886 932 251430, Fax: +886 2 600-0132 On Tue, 12 Aug 1997, Tom Minchin wrote: > Easily fixed. Just deny access to pictographics.com to your users. A more > rational approach (in face of a frothing webmaster) is to use cache_stoplist > on his domain. Then your users will only be proxying his pages, not caching > them. > > Alternatively, since you're in Belgium and he's in the US is to ignore him. > > tom@interact.net.au > > > Hi all, > > > > I got the attached message (see below) in my mailbox. > > "dalet.belnet.be" is our network wide parent cache. Although I > > don't think the request can be enforced, I'd like your opinions about > > this and if possible pointers to 'official' documents commenting on this > > issue. > > > > Thanks, > > > > -- > > Jan Torreele E-mail: Jan.Torreele@belnet.be > > BELNET Service Support Team Tel.: +32 (0) 2 / 238 34 70 > > > > ---------- Forwarded message ---------- > > Date: Sun, 10 Aug 1997 09:37:24 +0100 > > From:webmaster@pictographics.com > > To: sst@belnet.be > > Subject: caching of files > > > > Hello, > > > > I noticed this morning that your server "dalet.belnet.be" was caching every > > files on my server. While I appreciate your interest in these files, you > > will note from the terms and conditions listed on every single homepage on > > my website. > > > > Therefore, while I have no problem with your users visiting one or more of > > the pages served from my server, I DO NOT GIVE YOU PERMISSION to store > > these files on your server. > > > > I INSIST THAT YOU STOP COPYING FILES FROM MY SERVER IMMEDIATELY. I ALSO > > INSIST THAT YOU DELETE ALL FILES CURRENTLY STORED ON YOUR SERVER > > IMMEDIATELY. > > > > FAILURE TO DO SO WILL RESULT IN LEGAL ACTION. > > > > Thank you. > > > > Sincerely, > > > > Luke Dalla Bona > > President > > > > > > --------------------------------------------- > > Pictographics, Ltd. Serving and Designing Web Sites > > 69 Lansdowne Ave > > Sault Ste. Marie, Ontario P6B 1K5 > > 1.705.949.4871 (phone/fax) > > http://www.pictographics.com > > > > > Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA04551; Tue, 12 Aug 1997 07:01:52 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 07:01:52 -0700 (PDT) Message-Id: <3.0.32.19970812145938.006a7140@popmail.i-way.co.uk> X-Sender: neville@popmail.i-way.co.uk X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Tue, 12 Aug 1997 14:59:38 +0100 To: squid-users@nlanr.net From: Neville Stott <neville@i-way.net.uk> Subject: squid.conf Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"GQI6kS-bi22.0.QR6.Go6yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/199 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hello there, I am sure this has been asked but is there a squid.conf explanation guide as there are so many variables not all been obvious to me. Thanks in anticipation Neville Stott Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA21939; Tue, 12 Aug 1997 07:24:34 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 07:24:34 -0700 (PDT) X-Sender: gleeson@pixel.its.unimelb.edu.au Message-Id: <v03102801b0161f50c1cc@[10.0.2.15]> In-Reply-To: <Pine.3.89.9708121444.A14617-0100000@mahler.belnet.be> References: <199708120902.TAA16515@grey.interact.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 13 Aug 1997 00:21:27 +1000 To: Jan Torreele <Jan.Torreele@belnet.be> From: Martin Gleeson <gleeson@unimelb.edu.au> Subject: Re: Caching violates copyright? Cc: squid-users@nlanr.net Resent-Message-ID: <"84ilYQHMDqF.0.t93.I87yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/200 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Jan Torreele <Jan.Torreele@belnet.be> wrote: >On Tue, 12 Aug 1997, Tom Minchin wrote: >> Alternatively, since you're in Belgium and he's in the US is to ignore >> him. >:-) my thinking too. Indeed. I have heard that Netscape asked (in brutal legalese threats) one of Australia's largest ISPs to stop caching their sites. The ISP said "no". And that was it. Seriously, do you think it would be worth it to anyone to pay lawyers a fortune to sue someone in a different country on a very dubious basis just to get a few extra hits? >But the question really is, *is* there a copyright violation going >on here? The answer is probably not, but it hasn't been tested. You would expect that proxy servers would be covered under any "fair use" provision of copyright, but to date, I don't believe any legislation has been proposed or any cases tried. And even if it had, you've still got the age-old issue of international boundaries. So it's still up in the air (or on the wires, if you like :-). Cheers, Marty. ------------------------------------------------------------------------- Martin Gleeson Webmeister | http://www.unimelb.edu.au/%7Egleeson/ Information Technology Services | Email : gleeson@unimelb.edu.au The University of Melbourne, Oz. | Opinions : Mine, all mine. Imagine, if you will, a world without hypothetical situations... ------------------------------------------------------------------------- Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA21702; Tue, 12 Aug 1997 07:51:29 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 07:51:29 -0700 (PDT) Date: Tue, 12 Aug 1997 22:50:04 +0800 (PST) From: "Francis A. Vidal" <francis@usls.edu> To: Squid Users Mailing List <squid-users@nlanr.net> Subject: denying access to specific files Message-ID: <Pine.LNX.3.96.970812224639.3503C-100000@linux1.usls.edu> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"FuimYuz9FP5.0.Lb6.NX7yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/201 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net hello everyone! i wanted to prevent users from downloading files like .wav .viv .mov .mpeg .mpg, etc. because the workstations they're using doesn't have multi-media capabilities and it's just a waste of bandwidth if some curious kids download these types of files. how do i do it from the ACL entries? thanks! -- francis vidal [ systems administrator ] USLSnet, univ. of st. la salle la salle avenue, bacolod city http://www.usls.edu +63(34) 435-2324 P G P key available at - - ftp://ftp.usls.edu/pub/pgpkey/francis.pgp Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA30039; Tue, 12 Aug 1997 08:18:29 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 08:18:29 -0700 (PDT) Message-ID: <33F07E1A.71E0@cruiser.com.br> Date: Tue, 12 Aug 1997 12:15:39 -0300 From: "Vinicius D 'Antonio" <vda@cruiser.com.br> X-Mailer: Mozilla 3.01 (Win95; I) MIME-Version: 1.0 To: Neville Stott <neville@i-way.net.uk> CC: squid-users@nlanr.net Subject: Re: squid.conf References: <3.0.32.19970812145938.006a7140@popmail.i-way.co.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"yuMM0aavWB7.0.UF5.uv7yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/202 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, take a look at QUICKSTART in the your squid directory regards Neville Stott wrote: > > Hello there, > > I am sure this has been asked but is there a squid.conf explanation guide > as there are so many variables not all been obvious to me. > > Thanks in anticipation > > Neville Stott Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA08733; Tue, 12 Aug 1997 08:25:18 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 08:25:18 -0700 (PDT) Message-Id: <3.0.2.32.19970812161941.00c741bc@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Tue, 12 Aug 1997 16:19:41 +0100 To: squid-users@nlanr.net From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Documentation of src/Makefile.in Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"e7ncDMbqkQ3.0.e95.H18yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/203 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Could someone take the trouble of documenting all the options that can be defined in src/Makefile.in in the FAQ (and/or Oskar's user guide). I would do it myself, but the point is that I don't know what they all do!!! Volunteers? I wouldn't have thought it would take long - just a short paragraph per option. Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA11601; Tue, 12 Aug 1997 08:40:12 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 08:40:12 -0700 (PDT) From: "Andreas Strotmann" <Strotmann@rrz.Uni-Koeln.DE> Message-Id: <9708121739.ZM58921@rs3.rrz.Uni-Koeln.DE> Date: Tue, 12 Aug 1997 17:39:20 +0200 In-Reply-To: Jan Torreele <Jan.Torreele@belnet.be> "Caching violates copyright?" (Aug 12, 14:10) References: <Pine.3.89.9708121013.A10496-0100000@mahler.belnet.be> X-Mailer: Z-Mail (3.2.1 10oct95) To: Squid Users <squid-users@nlanr.net> Subject: Re: Caching violates copyright? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Resent-Message-ID: <"Ke_x5JeZWT7.0.J56.5F8yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/204 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, not in Germany it doesn't, and by inference probably neither does it in the rest of the European Union, including Belgium. In Germany there's a new law currently going through the legislative process that's relevant here. A couple of months ago I read a draft version; as far as I know, that draft is still pretty much what the end version is going to look like. As far as I remember, I was relieved to note that according to that draft law, caching (described as automatic, user-demand driven(!) store-and-forward mechanisms) is explicitly recognized *not* to violate any copyright at all! Note the "user-demand driven", though: pre-fetching probably _would_ constitute a violation! The central idea is that caching is just part of the transmission process, as some people on this list have been pointing out, "a very long line", and so is its legal status. But in order for that to be its legal status, it has to fit the discription: purely automatic (no human in the loop except to set up and manage the general setup), purely "reactive" (i.e. no pre-fetch). Another even more important aspect of that draftlaw from our point of view is that we (running the cache) are explicitly not liable for any contents of that cache (unless of course we ourselves caused it to be in there as users): as an example, just owning child pornography is a federal offense punishable by many years in prison around here (good thing, too) -- but fortunately we're not tainted by a user who downloads such stuff via our proxy server (which stores it, after all), much to our relief. Such a user, of course, does face a prison sentence if he keeps the stuff around. Despite this being in another country entirely, Belgium is part of the same European Union -- so if push came to shove, it shouldn't be too difficult to make a case that this German law does set a precedent in its neighbor country. (Oh, and this is NOT legal advice, of course ;-) Best, Andreas -- Andreas Strotmann / ~~~~~~ \________________A.Strotmann@Uni-Koeln.DE Universitaet zu Koeln /| University of Cologne \ Regionales Rechenzentrum| Regional Computer Center \ Robert-Koch-Str. 10 /| Tel: +49-221-478-5524 |\ Home: -221-4200663 D-50931 Koeln __|__ FAX: +49-221-478-5590 |__________~~~~~~~~~~~~ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA20123; Tue, 12 Aug 1997 09:32:20 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 09:32:20 -0700 (PDT) Message-Id: <199708121625.JAA21914@nlanr.net> To: squid-users@nlanr.net Subject: Re: Documentation of src/Makefile.in In-reply-to: Your message of Tue, 12 Aug 1997 16:19:41 +0100 Date: Tue, 12 Aug 97 09:25:28 -0700 From: Duane Wessels <wessels> X-Mts: smtp Resent-Message-ID: <"Kc2CJDC0CW4.0.nI7.xv8yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/205 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net JLarmour@origin-at.co.uk writes: >Could someone take the trouble of documenting all the options that can be >defined in src/Makefile.in in the FAQ (and/or Oskar's user guide). I would >do it myself, but the point is that I don't know what they all do!!! Done. http://squid.nlanr.net/Squid/FAQ/FAQ-3.html#ss3.1 Duane W. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA07957; Tue, 12 Aug 1997 09:49:48 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 09:49:48 -0700 (PDT) From: fsmith@spec.com (Frank Smith) Message-Id: <9708121643.AA23874@ss1000> Subject: Multiple cache_dir configuration ? To: squid-users@nlanr.net Date: Tue, 12 Aug 1997 11:43:45 -0500 (CDT) X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Resent-Message-ID: <"-X03RCcw9b5.0.GU7.nE9yp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/206 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net I'm trying to have squid use multiple cache directories. The config file claims you can use multiple cache_dir lines to spread out your cache, but there is only one cache_swap setting to set the maximum size. I would like to be able to set a size for each directory. If the cache_swap is set to be greater than the space on the first disk, will it automatically use the second, crash, or use equal amonts on both? Is it possible to control how big each cache_swap is, and if so, how? Thanks, frank -- Frank Smith -- Network Administrator E-mail: fsmith@spec.com Systems & Processes Engineering Corp. (SPEC) Voice:(512) 306-1100 x2154 401 Camp Craft Road Fax: (512) 306-1122 Austin, TX 78746-6558 Web: http://www.spec.com Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id KAA21293; Tue, 12 Aug 1997 10:50:49 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 10:50:49 -0700 (PDT) Date: Tue, 12 Aug 1997 12:09:28 -0500 (CDT) From: Matt Isleb <kermit@reptile.clc.cc.il.us> To: squid-users@nlanr.net Subject: Re: squid.conf In-Reply-To: <3.0.32.19970812145938.006a7140@popmail.i-way.co.uk> Message-ID: <Pine.LNX.3.91.970812120755.1187B-100000@reptile.clc.cc.il.us> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"mTFHTIzek2F.0.vl7.I7Ayp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/207 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Neville Stott wrote: > I am sure this has been asked but is there a squid.conf explanation guide > as there are so many variables not all been obvious to me. Isn't the default squid.conf commented well enough? And if the comments are not sufficient.. check out the Release-1.1 notes in the source tree. Matt Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id LAA30637; Tue, 12 Aug 1997 11:00:56 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 11:00:56 -0700 (PDT) Date: Tue, 12 Aug 1997 12:19:28 -0500 (CDT) From: Matt Isleb <kermit@reptile.clc.cc.il.us> To: "Francis A. Vidal" <francis@usls.edu> cc: Squid Users Mailing List <squid-users@nlanr.net> Subject: Re: denying access to specific files In-Reply-To: <Pine.LNX.3.96.970812224639.3503C-100000@linux1.usls.edu> Message-ID: <Pine.LNX.3.91.970812121155.1187C-100000@reptile.clc.cc.il.us> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"8YC_PT3sbQC.0.It3.mHAyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/208 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Francis A. Vidal wrote: > i wanted to prevent users from downloading files like .wav .viv .mov .mpeg > .mpg, etc. because the workstations they're using doesn't have multi-media > capabilities and it's just a waste of bandwidth if some curious kids > download these types of files. You administrative fascist! :-) I am also a bit of a fascist here. We don't like ppl accessing web chat becasuse they spend hours on it and there are ppl that need the lab computers to do homework. Create acl's like this: acl nochat src 207.57.233.128/27 207.57.239.0/24 ... acl chatdot url_regex http://chat.* http_access deny chatdot nochat (we have MANY more per site restrictions... hundreds even.) i.e. you might have "acl mpeg url_regex *.mpg" Hope that helps Matt Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id LAA24463; Tue, 12 Aug 1997 11:03:21 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 11:03:21 -0700 (PDT) Message-Id: <3.0.3.32.19970812140151.00c7c0fc@monroe.lib.mi.us> X-Sender: kip@monroe.lib.mi.us X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Tue, 12 Aug 1997 14:01:51 -0400 To: squid-users@nlanr.net From: Kip DeGraaf <kip@monroe.lib.mi.us> Subject: Re: Caching violates copyright? In-Reply-To: <3.0.3.32.19970812084918.00cb5da0@monroe.lib.mi.us> References: <199708121233.JAA01637@moose.webworks.ca> <Pine.3.89.9708121444.A14617-0100000@mahler.belnet.be> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"yhrak8bHCN9.0.Fe3.CKAyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/209 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net And the reply came back: >I have no problem with you caching files for your cache proxy. >Thank you kindly for asking. I have a simple question that perhaps you can >help me out with. If a page should change at my end, after it has been >cached on a proxy server, how is the change reflected in the cached version? Sounds like he was worried about people not getting the most up-to-date sales info. At 08:49 AM 8/12/97 -0400, Kip DeGraaf wrote: >Already have, just waiting for a reply. > >At 09:33 AM 8/12/97 -0300, Michael 'Moose' Dinn wrote: >>Why not mail the folks who are running the legality-laden web site and see >>what their policy is on caches? Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id LAA13282; Tue, 12 Aug 1997 11:36:46 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 11:36:46 -0700 (PDT) Date: Tue, 12 Aug 1997 14:34:05 -0400 (EDT) From: Michael Pelletier <mikep@comshare.com> To: Kip DeGraaf <kip@monroe.lib.mi.us> cc: squid-users@nlanr.net Subject: Re: Caching violates copyright? In-Reply-To: <3.0.3.32.19970812140151.00c7c0fc@monroe.lib.mi.us> Message-ID: <Pine.BSI.3.95.970812143253.494r-100000@valhalla.comshare.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"SnOBPCjMAF3.0.ep7.6pAyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/210 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Kip DeGraaf wrote: > And the reply came back: > > >I have no problem with you caching files for your cache proxy. > > >Thank you kindly for asking. I have a simple question that perhaps you can > >help me out with. If a page should change at my end, after it has been > >cached on a proxy server, how is the change reflected in the cached version? > > Sounds like he was worried about people not getting the most up-to-date > sales info. This seems to be a pretty common concern from people who don't understand a lot about web caching. Perhaps a explanatory page could be added to the Squid web page, going over the miscellany of Expires: headers and If-Modified-Since requests? -Mike Pelletier. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id LAA29417; Tue, 12 Aug 1997 11:40:18 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 11:40:18 -0700 (PDT) Message-ID: <33F0AD1E.5511B1B3@hem.passagen.se> Date: Tue, 12 Aug 1997 20:36:14 +0200 X-Mailer: Mozilla 3.01Gold (X11; I; Linux 2.0.30 i586) MIME-Version: 1.0 Subject: Re: Multiple cache_dir configuration ? References: <9708121643.AA23874@ss1000> Content-Type: text/plain; charset=iso-8859-1 To: Frank Smith <fsmith@spec.com> From: Henrik Nordstrom <hno@hem.passagen.se> Sender: henrik@pchno.memory.se Cc: squid-users@nlanr.net Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by nlanr.net id LAA01161 Resent-Message-ID: <"D__msrqDLmD.0.8a6.qsAyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/211 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Isn't this in the FAQ?? Ecah cache_dir has to be of equal size. It you have two disks where one is twice the size of the other you have to divide the first disk in two cache_dir's (not partitions..) --- Henrik Nordström Frank Smith wrote: > > I'm trying to have squid use multiple cache directories. The config > file claims you can use multiple cache_dir lines to spread out your > cache, but there is only one cache_swap setting to set the maximum > size. I would like to be able to set a size for each directory. If > the cache_swap is set to be greater than the space on the first > disk, will it automatically use the second, crash, or use equal amonts > on both? > Is it possible to control how big each cache_swap is, and if so, how? > > Thanks, > frank > > -- > Frank Smith -- Network Administrator E-mail: fsmith@spec.com > Systems & Processes Engineering Corp. (SPEC) Voice:(512) 306-1100 x2154 > 401 Camp Craft Road Fax: (512) 306-1122 > Austin, TX 78746-6558 Web: http://www.spec.com Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id LAA20166; Tue, 12 Aug 1997 11:42:02 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 11:42:02 -0700 (PDT) Date: Wed, 13 Aug 1997 02:39:34 +0800 (PST) From: "Francis A. Vidal" <francis@usls.edu> To: Matt Isleb <kermit@reptile.clc.cc.il.us> cc: Squid Users Mailing List <squid-users@nlanr.net> Subject: Re: denying access to specific files In-Reply-To: <Pine.LNX.3.91.970812121155.1187C-100000@reptile.clc.cc.il.us> Message-ID: <Pine.LNX.3.96.970813023331.5395A-100000@linux1.usls.edu> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"Yf0VmUZFsT5.0.VM7.QuAyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/212 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Matt Isleb wrote: > You administrative fascist! :-) > > I am also a bit of a fascist here. We don't like ppl accessing web chat > becasuse they spend hours on it and there are ppl that need the lab > computers to do homework. Create acl's like this: yep. my point exactly. students wasting time downloading files which they cannot play on weak machines. > acl nochat src 207.57.233.128/27 207.57.239.0/24 > > ... > > acl chatdot url_regex http://chat.* > http_access deny chatdot nochat hmmmmm... what about: acl files-denied urlpath_regex "/path/to/file/files-denied" http_access allow my-clients !files-denied /path/to/file/files-denied -------------------------- *.mpg *.mpeg *.vid *.viv *.wav > (we have MANY more per site restrictions... hundreds even.) maybe you can share them with me, if you don't mind at all =) -- francis vidal [ systems administrator ] USLSnet, univ. of st. la salle la salle avenue, bacolod city http://www.usls.edu +63(34) 435-2324 P G P key available at - - ftp://ftp.usls.edu/pub/pgpkey/francis.pgp Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id OAA14185; Tue, 12 Aug 1997 14:02:07 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 14:02:07 -0700 (PDT) Date: Wed, 13 Aug 1997 06:53:23 +1000 (EST) From: David J N Begley <david@avarice.nepean.uws.edu.au> Reply-To: David J N Begley <d.begley@nepean.uws.edu.au> To: Squid Users <squid-users@nlanr.net> Subject: Re: Caching violates copyright? In-Reply-To: <Pine.BSI.3.95.970812143253.494r-100000@valhalla.comshare.com> Message-ID: <Pine.LNX.3.96.970813065241.13528T-100000@avarice.nepean.uws.edu.au> X-No-Archive: Yes MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"wdLneRhXDFA.0.eX7.itCyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/213 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Michael Pelletier wrote: > This seems to be a pretty common concern from people who don't understand > a lot about web caching. Perhaps a explanatory page could be added to the > Squid web page, going over the miscellany of Expires: headers and > If-Modified-Since requests? Don't overdo it, though - we don't won't half the Web to suddenly be uncacheable. ;-) dave Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id OAA29577; Tue, 12 Aug 1997 14:43:19 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 14:43:19 -0700 (PDT) Message-ID: <B0000098933@genesis.wisper.net> From: "Leigh Porter" <leigh@wisper.net> To: "David J N Begley" <d.begley@nepean.uws.edu.au>, "Squid Users" <squid-users@nlanr.net> Subject: Re: Caching violates copyright? Date: Thu, 10 Apr 1997 22:25:20 +0100 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1157 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"td8zEoolhoA.0.4F7.7YDyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/214 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net ---------- > From: David J N Begley <david@avarice.nepean.uws.edu.au> > To: Squid Users <squid-users@nlanr.net> > Subject: Re: Caching violates copyright? > Date: Tuesday, August 12, 1997 9:53 PM > > On Tue, 12 Aug 1997, Michael Pelletier wrote: > > Don't overdo it, though - we don't won't half the Web to suddenly be > uncacheable. ;-) That's OK, so long as it's the half that nobody looks at :) -- Leigh Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id QAA03199; Tue, 12 Aug 1997 16:57:45 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 16:57:45 -0700 (PDT) Date: Tue, 12 Aug 1997 18:54:32 -0400 (EDT) From: Gregory Maxwell <nullc@nightshade.ml.org> To: Henny Bekker <Henny.Bekker@cc.ruu.nl> cc: Jan Torreele <Jan.Torreele@belnet.be>, Tom Minchin <tom@interact.net.au>, squid-users@nlanr.net Subject: Re: Caching violates copyright? In-Reply-To: <3.0.1.32.19970812134906.0113c8f0@hydra.cc.ruu.nl> Message-ID: <Pine.LNX.3.95.970812185240.22282B-100000@nightshade.ml.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"Jf3_ld6CcDD.0.9N.mVFyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/215 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Tue, 12 Aug 1997, Henny Bekker wrote: > At 02:17 PM 8/12/97 +0200, Jan Torreele wrote: > > [...Stuff deleted...] > > >But the question really is, *is* there a copyright violation going > >on here? After all, he could also complain that there are illegal copies > >of is stuff sitting in the buffers of my routers, etc. In my opinion I'm > >just providing an infrastructure that allow my users to efficiently > >access documents publicly accessible on the Internet. > > > In my opiniom a caching-server (in this case a Web caching-server but it > also applies to other caching such as ordinary disk-caching servers) can > be considered as a (very) long network cable.. > > Cheers, Henny > > NB This applies _not_ for mirror servers.. > Yes, could we have someone publish: "Web Cache: A device that behaves like an infinitly long network cable with the ablity to re-load the last sever hundred megabytes sent at any time for performance reasons" :) Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id XAA19291; Tue, 12 Aug 1997 23:07:19 -0700 (PDT) Resent-Date: Tue, 12 Aug 1997 23:07:19 -0700 (PDT) Message-Id: <199708130603.XAA29315@f32.hotmail.com> X-Originating-IP: [203.120.194.199] From: "l w" <lxyzj@hotmail.com> To: squid-users@nlanr.net Subject: Re: get "Operation already in progress" when trying to use squid Content-Type: text/plain Date: Tue, 12 Aug 1997 23:03:40 PDT Resent-Message-ID: <"J9L2OL4odhD.0.NC2.jvKyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/216 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, I have the same proble. Could anybody tell me how to socksify the squid with socks5. thanks ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id AAA10157; Wed, 13 Aug 1997 00:13:26 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 00:13:26 -0700 (PDT) From: "A. Rytsola" <raccoon@newt.dlc.fi> Message-Id: <9708131013.ZM19120@newt.dlc.fi> Date: Wed, 13 Aug 1997 10:13:34 -0400 In-Reply-To: David J N Begley <david@avarice.nepean.uws.edu.au> "Re: Caching violates copyright?" (Aug 13, 6:53am) References: <Pine.LNX.3.96.970813065241.13528T-100000@avarice.nepean.uws.edu.au> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Squid Users <squid-users@nlanr.net> Subject: Re: Caching violates copyright? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Resent-Message-ID: <"uDmOIit5wM9.0.p13.OuLyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/217 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net > > Squid web page, going over the miscellany of Expires: headers and > > If-Modified-Since requests? > > Don't overdo it, though - we don't won't half the Web to suddenly be > uncacheable. ;-) I agree. M$ is bad enough. Antti Rytsola Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id BAA12634; Wed, 13 Aug 1997 01:19:58 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 01:19:58 -0700 (PDT) Date: Wed, 13 Aug 1997 10:16:34 +0200 (MET DST) From: Jan Torreele <Jan.Torreele@belnet.be> Sender: Jan Torreele <Jan.Torreele@belnet.be> Reply-To: Jan Torreele <Jan.Torreele@belnet.be> Subject: Re: Update: Caching violates copyright To: Squid Users <squid-users@nlanr.net> Message-ID: <Pine.3.89.9708131005.A14617-0100000@mahler.belnet.be> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Resent-Message-ID: <"LnFzckaFZT3.0.1R.JsMyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/218 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Seems like the discussion did some good after all. Below is a new message from the people at pictographics. I wonder whether the 'discussion with my legal people' refers to an email conversation with Kip DeGraaf... :-) CU, -- Jan Torreele E-mail: Jan.Torreele@belnet.be BELNET Service Support Team Tel.: +32 (0) 2 / 238 34 70 ---------- Forwarded message ---------- Date: Tue, 12 Aug 1997 14:57:03 +0100 From:webmaster@pictographics.com To: Jan Torreele <Jan.Torreele@belnet.be> Subject: Re: caching of files Dear Jan, I have looked into proxy server issues further and have discussed it over with my legal people. They suggest that proxy servers pose no problem or conflict with the terms and conditions as stated by Pictographics, Ltd. Apparently I was a little hasty sending out my email of Sunday morning. As clarification, please disregard my previous two emails and cache away!! Thanks in advance. Luke Dalla Bona --------------------------------------------- Pictographics, Ltd. Serving and Designing Web Sites 69 Lansdowne Ave Sault Ste. Marie, Ontario P6B 1K5 1.705.949.4871 (phone/fax) http://www.pictographics.com Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id EAA00207; Wed, 13 Aug 1997 04:05:18 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 04:05:18 -0700 (PDT) Date: 13 Aug 1997 12:15:00 +0200 From: ar@andromeda.netaccess.de (Andreas Reeh) To: squid-users@nlanr.net Message-ID: <6blRa-w$4lB@andromeda.netaccess.de> Subject: ERR_FTP_DISABLED X-Mailer: CrossPoint v3.11 R/C464 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Organization: NetAccess Datendienste GmbH, http://www.netaccess.de/ or info@netaccess.de Reply-To: webadmin@andromeda.netaccess.de Resent-Message-ID: <"whjIeRLFu6B.0.gB1.YFPyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/219 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi Squiders, since I upgraded from 1.NOVM.14 to 1.NOVM.15 I often see >Ready to serve requests. >97/08/13 12:07:09| Restarting ftpget server... >97/08/13 12:07:09| ftpget server failing too rapidly >97/08/13 12:07:09| WARNING: FTP access is disabled! >97/08/13 12:07:25| ERR_FTP_DISABLED: ftp://ftp.foo.bar/pub/win95/ ... in cache.log after I send Squid a "kill -HUP". This happens on to different machines, both are Linux 2.0.30. Andreas -- Andreas Reeh = A.Reeh@WaD.org (private) or ar@netaccess.de (business) More info, public key, geek-code and other email and talk addresses at http://www.wad.org/~andreas/ or http://home.pages.de/~andreas.reeh/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA10742; Wed, 13 Aug 1997 05:29:46 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 05:29:46 -0700 (PDT) Message-ID: <19970813142437.40953@admin.is.co.za> Date: Wed, 13 Aug 1997 14:24:37 +0200 From: Oskar Pearson <oskar@is.co.za> To: Ronald Wiplinger <ronald@linkou.trace.com.tw> Cc: Tom Minchin <tom@interact.net.au>, Jan Torreele <Jan.Torreele@belnet.be>, squid-users@nlanr.net Subject: Re: Caching violates copyright? References: <199708120902.TAA16515@grey.interact.net.au> <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69e In-Reply-To: <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw>; from Ronald Wiplinger on Tue, Aug 12, 1997 at 09:21:33PM +0800 Reply-To: oskar@is.co.za Resent-Message-ID: <"uAR08pRjWd3.0.Xb1.mXQyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/220 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi > I think the site should do something, if he don't like to cache, he can > expire the page, like M$ it do. >>> I noticed this morning that your server "dalet.belnet.be" was caching every >>> files on my server. While I appreciate your interest in these files, you How about this. Every person that is using squid or any other cache (or even if you are using netscape with the built-in disk cache turned on) send him a (not 5000...) copy of this message with the appropriate stuff changed. I just did. Let him know that he should get a clue before he goes crazy. ----------- To: webmaster@pictographics.com We (cache.is.co.za) also cache your web pages. If you do not wish them to be saved here feel free to sue us. If there are too many of us, perhaps you would think about trying to use the STANDARDS, which our software obeys, so that we don't cache pages. It's a little akin to putting a digital 'Copyright' signature on a page... but something that machines can read. ftp://ftp.isi.edu/in-notes/rfc/rfc2068.txt ftp.isi.edu:/in-notes/rfc/ Oskar ----------- Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA02251; Wed, 13 Aug 1997 05:47:48 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 05:47:48 -0700 (PDT) Message-Id: <3.0.3.32.19970813085013.03784250@monroe.lib.mi.us> X-Sender: kip@monroe.lib.mi.us X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Wed, 13 Aug 1997 08:50:13 -0400 To: squid-users@nlanr.net From: Kip DeGraaf <kip@monroe.lib.mi.us> Subject: Re: Caching violates copyright? In-Reply-To: <19970813142437.40953@admin.is.co.za> References: <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> <199708120902.TAA16515@grey.interact.net.au> <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"FC2WV888374.0.MG7.ipQyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/221 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Okay, wait a second. Pictographics has gotten the message and understands what a caching proxy really does. Let's not alienate the gentleman in the process. Look at the last messages from myself and Jan Torreele, we've both had reasonable and pleasant exchanges with the site. (Of course, if he hadn't been reasonable, this sounds like an interesting idea) At 02:24 PM 8/13/97 +0200, Oskar Pearson wrote: >How about this. Every person that is using squid or any other cache >(or even if you are using netscape with the built-in disk cache turned on) >send him a (not 5000...) copy of this message with the appropriate stuff >changed. I just did. Let him know that he should get a clue before >he goes crazy. > >----------- >To: webmaster@pictographics.com > >We (cache.is.co.za) also cache your web pages. If you do not wish them to >be saved here feel free to sue us. If there are too many of us, perhaps you >would think about trying to use the STANDARDS, which our software obeys, >so that we don't cache pages. It's a little akin to putting a digital >'Copyright' signature on a page... but something that machines can read. > >ftp://ftp.isi.edu/in-notes/rfc/rfc2068.txt >ftp.isi.edu:/in-notes/rfc/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA15327; Wed, 13 Aug 1997 07:48:47 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 07:48:47 -0700 (PDT) Date: 13 Aug 1997 14:01:00 +0200 From: ar@andromeda.netaccess.de (Andreas Reeh) To: squid-users@nlanr.net Message-ID: <6blRj9w$4lB@andromeda.netaccess.de> Subject: Meaning of "Page faults with physical IO" ? X-Mailer: CrossPoint v3.11 R/C464 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Organization: NetAccess Datendienste GmbH, http://www.netaccess.de/ or info@netaccess.de Resent-Message-ID: <"m5AH2K9kQ0B.0.Xv2.XYSyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/222 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi Squiders, what does "Page faults with physical i/o: 4897" in my Cachemgr "Cache Information"-output mean ? Andreas -- Andreas Reeh = A.Reeh@WaD.org (private) or ar@netaccess.de (business) More info, public key, geek-code and other email and talk addresses at http://www.wad.org/~andreas/ or http://home.pages.de/~andreas.reeh/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA17367; Wed, 13 Aug 1997 08:20:52 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 08:20:52 -0700 (PDT) Date: Wed, 13 Aug 1997 17:18:03 +0200 From: Paola Tentoni <tentoni@cilea.it> Subject: Bus error ... on HP-UX running squid 1.1.14 X-Sender: tentoni@icil64.cilea.it To: squid-users@nlanr.net Message-id: <2.2.32.19970813151803.01022d20@icil64.cilea.it> MIME-version: 1.0 X-Mailer: Windows Eudora Pro Version 2.2 (32) Content-type: text/plain; charset="us-ascii" Resent-Message-ID: <"3zgxS39MEb8.0.r13.X1Typ"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/223 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, I'am trying to upgrade my squid 1.1.0 on a HP-UX 9.05 to the 1.1.15 (or 14) version. I have a lot of problems, as it dies almost immediatly: 97/08/13 16:30:12| Starting Squid Cache version 1.1.15 for hppa1.1-hp-hpux9.05... 97/08/13 16:30:12| With 2048 file descriptors available 97/08/13 16:30:12| Performing DNS Tests... 97/08/13 16:30:12| Successful DNS name lookup tests... 97/08/13 16:30:12| Started 15 'dnsserver' processes 97/08/13 16:30:12| Unlinkd pipe opened on FD 22 97/08/13 16:30:12| file_map_create: creating space for 2097152 files 97/08/13 16:30:12| Swap maxSize 102400, estimated 5120 objects 97/08/13 16:30:12| Target number of buckets: 256 97/08/13 16:30:12| Using 7951 Store buckets, maintain 1 bucket every 10 seconds 97/08/13 16:30:12| Store logging disabled 97/08/13 16:30:12| Rebuilding storage from disk image in /usr/local/squidt/cache 97/08/13 16:30:14| Accepting HTTP connections on FD 25. 97/08/13 16:30:14| Accepting ICP connections on FD 26. 97/08/13 16:30:14| Pinger socket opened on FD 27 97/08/13 16:30:14| Ready to serve requests. 97/08/13 16:30:14| Finished rebuilding storage from disk image. 97/08/13 16:30:14| 0 Lines read from previous logfile. 97/08/13 16:30:14| 0 Objects loaded. 97/08/13 16:30:14| 0 Objects expired. 97/08/13 16:30:14| 0 Duplicate URLs purged. 97/08/13 16:30:14| 0 Swapfile clashes avoided. 97/08/13 16:30:14| Took 0 seconds ( 0.0 objects/sec). 97/08/13 16:30:14| store_swap_size = 0k 97/08/13 16:30:14| ICMP socket opened 97/08/13 16:30:14| Configuring Parent eureka.lucia.it/8080/3130 97/08/13 16:30:14| Configuring Sibling proxy.cineca.it/8080/3130 FATAL: Received Bus Error...dying. 97/08/13 16:31:59| storeWriteCleanLog: Starting... 97/08/13 16:31:59| Finished. Wrote 3 lines. 97/08/13 16:31:59| Took 0 seconds ( 3.0 lines/sec). CPU Usage: user 0 sys 0 Maximum Resident Size: 1756 KB Page faults with physical i/o: 12 Memory usage for squid via mallinfo(): total space in arena: 0 KB Ordinary blocks: 0 KB 0 blks Small blocks: 0 KB 0 blks Holding blocks: 0 KB 0 blks Free Small blocks: 0 KB Free Ordinary blocks: 0 KB Total in use: 0 KB 0% Total free: 0 KB 0% 97/08/13 16:32:18| Pinger exiting. The problems is quite the same with version 1.1.14 (and also 1.1.11). It caches some URL (two or three, written in the log) than it dies. Compiler used: gcc.2.7.2.1. Any help will be quite appreciated, Paola Tentoni * Paola Tentoni * CILEA - Via R.Sanzio 4, Segrate (Milan) Italy * tel +39 2 26922434 FAX: +39 2 2135520 * e-mail: tentoni@cilea.it Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA13631; Wed, 13 Aug 1997 09:19:04 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 09:19:04 -0700 (PDT) Sender: eeindy@eei.ericsson.se Message-ID: <33F1DD7F.F7EBE55A@eei.ericsson.se> Date: Wed, 13 Aug 1997 17:14:55 +0100 From: Niall Doherty <eeindy@eei.ericsson.se> Organization: Ericsson Systems Expertise Limited X-Mailer: Mozilla 4.02 [en] (X11; I; SunOS 5.5.1 sun4u) MIME-Version: 1.0 To: squid-users@nlanr.net Subject: Accept Failure: (71) Protocol error Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"RzCSzqGC2JC.0.Eq2.ssTyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/224 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, The cache log is showing quite a few of these errors - between 40-80 each day. 97/08/13 16:35:38| comm_accept: FD 50: accept failure: (71) Protocol error 97/08/13 16:35:38| asciiHandleConn: FD 50: accept failure: (71) Protocol error 97/08/13 16:35:38| comm_accept: FD 50: accept failure: (71) Protocol error 97/08/13 16:35:38| asciiHandleConn: FD 50: accept failure: (71) Protocol error FD 50 is the HTTP socket. Anyone got any idea as to what's causing these ? UltraSPARC, Solaris 2.5.1, Squid 1.1.15 (occurred with 1.1.11 too). Browsers are mainly Netscape (version 2->4) and MSIE. Thanks, Niall -- Niall Doherty Systems Engineer Ericsson Systems Expertise Ltd., Beech Hill, Telephone: +353 1 207 7506 Clonskeagh, Telefax: +353 1 269 3153 Dublin 4, Email: eeindy@eei.ericsson.se Ireland. MemoID: EEI.EEINDY ECN: 830 7506 Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA12317; Wed, 13 Aug 1997 09:45:20 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 09:45:20 -0700 (PDT) Message-Id: <3.0.2.32.19970813173301.00c6077c@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Wed, 13 Aug 1997 17:33:01 +0100 To: ar@andromeda.netaccess.de (Andreas Reeh), squid-users@nlanr.net From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Meaning of "Page faults with physical IO" ? Cc: wessels@nlanr.net In-Reply-To: <6blRj9w$4lB@andromeda.netaccess.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"HgIAFzP0jfE.0.mF1.LCUyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/225 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 14:01 13/08/97 +0200, Andreas Reeh wrote: >Hi Squiders, > >what does "Page faults with physical i/o: 4897" in my Cachemgr "Cache >Information"-output mean ? You get a "page fault", when your OS tries to access something in memory which is actually swapped to disk. The term "page fault", while correct at the kernel and CPU level, is a bit deceptive to a user, as there's no actual error - this is a normal feature of operation. Also, this doesn't necessarily mean your squid is swapping by that much. Most operating systems also implement paging for executables, so that only sections of the executable which are actually used are read from disk into memory. Also, whenever squid needs more memory, the fact that the memory was allocated will show up in the page faults. However, if the number of faults is unusually high, and getting bigger, this could mean that squid is swapping. Another way to verify this is using a program called "vmstat" which is found on most UNIX platforms. If you run this as "vmstat 5", this will update a display every 5 seconds. This can tell you if the system as a whole is swapping a lot (see your local man page for vmstat for more information). It is very bad for squid to swap, as every single request will be blocked until the requested data is swapped in. It is better to tweak the cache_mem and/or memory_pools setting in squid.conf, or switch to the NOVM versions of squid, than allow this to happen. [To Duane: Could this be added to the FAQ in the cache manager section, assuming no-one has any objections?] Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id RAA12813; Wed, 13 Aug 1997 17:14:02 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 17:14:02 -0700 (PDT) Message-Id: <s3f20c17.077@swmotor.usxpress.com> X-Mailer: Novell GroupWise 4.1 Date: Wed, 13 Aug 1997 20:15:34 -0400 From: Paul Gracy <GRACYP@swmotor.usxpress.com> To: squid-users@nlanr.net Subject: problem starting squid Mime-Version: 1.0 Content-Type: text/plain Content-Disposition: inline Resent-Message-ID: <"Lk02VMDMMbF.0.wP4.Cpayp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/226 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net I just subscribed to this list so please email directly to me at gracyp@usxpress.com. Thanks. I keep having problems starting squid. FATAL: Cannot open swap logfile: /usr/local/squid/cache/log FATAL: storeRebuiltFromDisk: rename failed I've tried various deletions, squid -z, etc. What am I missing? Paul Gracy Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id RAA24287; Wed, 13 Aug 1997 17:37:21 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 17:37:21 -0700 (PDT) From: Hwang Hyojung <jung@jean.kotel.co.kr> Message-Id: <199708140032.JAA11609@jean.kotel.co.kr> Subject: Re: problem starting squid To: GRACYP@swmotor.usxpress.com (Paul Gracy) Date: Thu, 14 Aug 1997 09:32:57 +0900 (KST) Cc: squid-users@nlanr.net In-Reply-To: <s3f20c17.077@swmotor.usxpress.com> from "Paul Gracy" at Aug 13, 97 08:15:34 pm X-Mailer: ELM [version 2.4 PL24 PGP2] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-2022-kr Content-Transfer-Encoding: 7bit Resent-Message-ID: <"ljlbCTcPgNA.0.LO3.OBbyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/227 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Maybe /usr/local/squid/cache directory's write permission is not allowed or owner might be wrong. Check this point first. According to Paul Gracy: > > I just subscribed to this list so please email directly to me at > gracyp@usxpress.com. Thanks. > > I keep having problems starting squid. > FATAL: Cannot open swap logfile: /usr/local/squid/cache/log > FATAL: storeRebuiltFromDisk: rename failed > > I've tried various deletions, squid -z, etc. What am I missing? > > Paul Gracy > > -- How astonishing it is to be alive today !! * _ _ ||_|| ||_| Hwang, Hyojung / jung@jean.kotel.co.kr / 82-2-526-5235 |_| _|_ Internet Team / Multimedia Laboratory / KT R&D Group |_| 137-792, 17 Umyun-Dong, Seocho-Gu, Seoul, Korea Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id TAA27424; Wed, 13 Aug 1997 19:50:57 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 19:50:57 -0700 (PDT) Message-Id: <199708140248.TAA21796@nlanr.net> To: Peter Marelas <maral@phase-one.com.au> cc: squid-users@nlanr.net Subject: Re: Thinking about ICP In-reply-to: Your message of Sat, 02 Aug 1997 20:02:33 +1000 Date: Wed, 13 Aug 97 19:48:25 -0700 From: Duane Wessels <wessels> X-Mts: smtp Resent-Message-ID: <"PEz_k4HP7cE.0.ci4.18dyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/228 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net [catching up on old email, sorry] maral@phase-one.com.au writes: >We currently have this problem at a National Exchange Point where we have >many siblings querying each other over multiple WAN's. > >The idea I put forward was to "cluster" the siblings in groups of 8 like so: > > > [ 8 cluser sibling ] [ 8 cluster sibling ] > x x > x x > x x > [ ICP relay ] > x x > x x > x x > [ 8 cluser sibling ] [ 8 cluster sibling ] > >All siblings in a cluster would also include the ICP relay system as a sibling. >The ICP relay system would fire off UDP_REQUESTS to all other siblings in >the "other" clusters. >The ICP relay would relay back to the original requestor a single MISS or a HIT with the real proxies h ostname so the object can be fetched. >What this elimintates is each sibling having to reply to the requestor with >a MISS. This is a really interesting idea, similar to what others have proposed or asked for. There is only one thing preventing this from being really easy to implement: Squid and ICP don't have any concept of third-party caches. At one point there was another cache package (mirror-image?) which was going to implment third-party "redirections" with ICP. There are a couple of ways the ICP relay could be implemented with some modifications/additions to ICP. If anyone is seriously interested in building one of these, please let me know and we can work it all out. Duane W. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id UAA18071; Wed, 13 Aug 1997 20:11:46 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 20:11:46 -0700 (PDT) Message-Id: <199708140310.UAA05438@nlanr.net> To: "Anthony J. Breeds-Taurima" <tony@cantech.net.au> cc: Squid Users <squid-users@nlanr.net> Subject: Re: Hmmm ... "Out of file descriptors" In-reply-to: Your message of Wed, 06 Aug 1997 16:45:36 +0800 Date: Wed, 13 Aug 97 20:10:09 -0700 From: Duane Wessels <wessels> X-Mts: smtp Resent-Message-ID: <"mWySxL6NzTE.0.Hv1.NSdyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/229 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net tony@cantech.net.au writes: >On Wed, 6 Aug 1997, Stuart Bryant wrote: > >> check your startup scripts to see if they dont start 2 copies of squid. >> or that something else doesnt start on that port. > >I definatly only have one copy of squid running. >As for the port, am I correct in reading >commBind: Cannot bind socket FD XXX to 203.26.6.6:0: (98) Address .... >as IPaddr: 203.26.6.6 >port: 0 ?? The port 0 is definately wrong. Which version are you running? Can you send more of the cache.log messages before this error? Duane W. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id UAA01972; Wed, 13 Aug 1997 20:39:42 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 20:39:42 -0700 (PDT) Message-Id: <199708140337.UAA16885@nlanr.net> To: andrew@ugh.net.au cc: squid-users@nlanr.net Subject: Re: Slow name lookups on non existant hosts In-reply-to: Your message of Thu, 07 Aug 1997 15:32:51 +1000 Date: Wed, 13 Aug 97 20:37:52 -0700 From: Duane Wessels <wessels> X-Mts: smtp Resent-Message-ID: <"_GbJ20DSFs8.0.Qp6.Tsdyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/230 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net andrew@ugh.net.au writes: >To solve this I need to get squid to lookup the host before passing the >query on...or perhaps I should make the parent a sibling and only get a >page if it already has it. This is probably better for us but worse for >all the other users of our parent proxy (as that proxy will ahve cached >less). I believe that you'd need to modify the source for this. Squid tries to be persistent and if the DNS lookup fails locally it forwards the request on to a parent which might have better luck getting the address resolved. Duane W. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id XAA05079; Wed, 13 Aug 1997 23:03:50 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 23:03:50 -0700 (PDT) Message-Id: <s3f2bb29.065@ml.petech.ac.za> X-Mailer: Novell GroupWise 4.1 Date: Thu, 14 Aug 1997 08:00:31 +0200 From: Bruce Smith <9401962@ml.petech.ac.za> To: squid-users@nlanr.net Subject: RE Help with setting non-caching of a site Mime-Version: 1.0 Content-Type: text/plain Content-Disposition: inline Resent-Message-ID: <"ciHfXSxUiW6.0.vM4.-yfyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/231 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi I'm a newbie to squid and need some advice. The users are having a problem with accessing one site. It is a online-workgroup area, but doesn't use things like cgi-bin or ? in the urls and the proxy is caching all the pages. This is a major problem and we need to sort it out. I'm not too sure of the extensions to the pages. Does anyone have any suggestions on this problem. Bruce Smith Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id XAA32125; Wed, 13 Aug 1997 23:39:38 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 23:39:38 -0700 (PDT) Message-Id: <3.0.1.32.19970814163716.00886360@email.aims.gov.au> X-Sender: kjones@email.aims.gov.au X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Thu, 14 Aug 1997 16:37:16 +1000 To: squid-users@nlanr.net From: Kerry Jones <kjones@aims.gov.au> Subject: Parents of Parent Proxies Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"v6rbyP0-Gk7.0.nP.1Vgyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/232 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, We have configured our Squid Proxy server as a sibling of another Proxy server and our Parent is currently using another Proxy server as its parent. Do we need to configure our Parent's Parent proxies as well as our Parent proxy or will a request from us to our Parent handle the requests to our Parent's Parent?? P.S. Did you understand the question?? -- Kerry Jones, UNIX System Manager __ __ Australian Institute of Marine Science, / `-' ) ,,, Townsville, Australia. | [====||||||||||[:::} k.jones@aims.gov.au \__.-._\ ``` Ph: (077) 534 400 ..................................................................... Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id XAA32720; Wed, 13 Aug 1997 23:55:10 -0700 (PDT) Resent-Date: Wed, 13 Aug 1997 23:55:10 -0700 (PDT) Date: Thu, 14 Aug 1997 16:42:52 +1000 (EST) From: David Richards <dj.richards@qut.edu.au> Subject: Re: RE Help with setting non-caching of a site In-reply-to: <s3f2bb29.065@ml.petech.ac.za> To: Bruce Smith <9401962@ml.petech.ac.za> Cc: squid-users@nlanr.net Message-id: <Pine.OSF.3.92.970814163758.9993B-100000@pigeon.qut.edu.au> MIME-version: 1.0 Content-type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"7LuJa--KB01.0.nH3.mjgyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/233 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Bruce, The tags "cache_stoplist" and "hierarchy_stoplist" will help you out. Just add in the domain or page name, or something else significant and that will fix it up for you. Example: If I had a page that was somewhere in the world that I know is being continually update, let's say: http://www.somewhere.com/page.html I would put in hierarchy_stoplist: hierarchy_stoplist cgi ? www.somewhere.com/page.html and in "cache_stoplist" cache_stoplist cgi ? www.somewhere.com/page.html That information is significant enough to tell squid that I do not want the page to be cached locally and do not want to fetch the page from a sibling or parent. It will then always get it from the source. Seeya, Dave. ---------------------------------------------------------------------- David Richards Ph: +61 7 3864 4354 Network Programmer Fax: +61 7 3864 5272 Computing Services e-mail: dj.richards@qut.edu.au Queensland University of Technology ---------------------------------------------------------------------- On Thu, 14 Aug 1997, Bruce Smith wrote: > Hi > > I'm a newbie to squid and need some advice. The users are having a > problem with accessing one site. It is a online-workgroup area, but > doesn't use things like cgi-bin or ? in the urls and the proxy is caching all > the pages. This is a major problem and we need to sort it out. I'm not too > sure of the extensions to the pages. > > Does anyone have any suggestions on this problem. > > Bruce Smith > Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id BAA24011; Thu, 14 Aug 1997 01:48:15 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 01:48:15 -0700 (PDT) Date: Thu, 14 Aug 1997 04:30:05 -0400 From: "Armistead, Jason" <ARMISTEJ@oeca.otis.com> Subject: Question about SSL using CONNECT method To: "'Squid-Users'" <squid-users@nlanr.net> Message-id: <c=US%a=_%p=UTC%l=UTC/OTUSIMC/000D0C15@otussimc.otis.com> X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.995.52 Encoding: 33 TEXT, 47 UUENCODE X-MS-Attachment: WINMAIL.DAT 0 00-00-1980 00:00 X-MS-TNEF-Correlator: <c=US%a=_%p=UTC%l=UTC/OTUSIMC/000D0C15@otussimc.otis.com> Resent-Message-ID: <"SGsZf0FFHTC.0.tx6.KMiyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/234 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi One of our Intranet sites is using HTTPS, which I have deduced is HTTP over SSL and hence non cacheable. Am I correct so far ? The site is using Basic authentication to gain entry to either a HTTP or a HTTPS directory tree (same information in both, but Basic authentication credentials causes IIS Active-Server pages change the menus in both too). I think they're really only worried about preventing orders taking place on their electronic shopping-basket parts inventory system. The rest is pretty plain old surfing as far as I have been able to determine (and doesn't really need HTTPS at all). Is there any way to make part of a site non SSL e.g. the GIF/JPEG images which are just bandwidth hogs, thus improve the caching, but to still be able to provide security on HTTP POST and GET requests (such as to the order system) ? OR, do they have to use two different realms so that the browser can work out which parts are SSL HTTPS requests, and which are plain old HTTP. Can such a mixed environment even exist ? At present for HTTPS Squid can only act as a PROXY, forwarding requests, and not a CACHE, storing the data, for SSL CONNECTs - am I right ? Regards Jason Armistead Systems Engineer OEC-Australia, Minto armistej@oeca.otis.com begin 600 WINMAIL.DAT M>)\^(@0(`0:0" `$```````!``$``0>0!@`(````Y 0```````#H``$(@ <` M& ```$E032Y-:6-R;W-O9G0@36%I;"Y.;W1E`#$(`0F `0`A````0C Y,T4P M,#$Y-C$T1#$Q,3E",#<P,# P1C@P,S4X,#$`L 8!(( #``X```#-!P@`#@`$ M`"@``@`$`!P!`06 `P`.````S0<(``X`! `>``4`! `5`0$-@ 0``@````(` M`@`!!( !`"@```!1=65S=&EO;B!A8F]U="!34TP@=7-I;F<@0T].3D5#5"!M M971H;V0`M@T!`Y &`+0&```7````0 `Y`"#RDD2,J+P!'@!P``$````H```` M475E<W1I;VX@86)O=70@4U-,('5S:6YG($-/3DY%0U0@;65T:&]D``(!<0`! M````%@````&\J(Q$D@'@D\,4EA'1FP<``/@#6 $```,`!A :E#EA`P`'$/4# M```>``@0`0```&4```!(24].14]&3U5224Y44D%.1513251%4TE355-)3D=( M5%104RQ72$E#2$E(059%1$5$54-%1$E32%144$]615)34TQ!3D1(14Y#14Y/ M3D-!0TA%04),14%-24-/4E)%0U133T9!``````,`$! ``````P`1$ `````" M`0D0`0```%8$``!2! ``:@8``$Q:1G5F4K*V_P`*`0\"%0*H!>L"@P!0`O() M`@!C: K <V5T,C<&``;#`H,R`\4"`'!R0G$1XG-T96T"@S.W`N0'$P*#- /& M%,@U$LQ=%,5]"H (SPG9.QC/,CPU-0* "H$-L0M@;F=X,3 S%% +"A11"_$@ M#$AI"H4*A4]N92!,;V8>H AP($D",')[`' 1P" `D!/0!" $`" .=0"0&] = M8%144%-@+"!W:&D1<!\0((T1@'8>D VP9'5C"8 ?( (@HAZ@(; %P%-33#8@ M`' B0&@)\"(@(&XY`B @8P#0([ !H&QE="X@%+!M(6$%H1C08U$?D6\@9@K! M/QV<5*<CL!^C( A"80"08R-@U'5T([%T(3!A*2 D(2IT)?!G"W$@*0%R>=TI MHF4?P".P!<!A(H4K%GD%\&1I)9(%L"IQ"=$@6"AS80> ( !N`A!ROP# *6,J M`0;@*. @\&(HT/\H7REC!0`)@"D"!T $("104R! !Y%)207P026P:=TAL"T& M82+R"K!G!Y$1<=<;T!Z0*.$@!X!N($ N5GDIH6\I).$A<"C@"X!K:3.2>2<8 MT" 8T = ;/<J< (@-G%W)7$(D") `:#W"& %0!-092&P*1$@<06P]P2!!" ! MD&L@8@M1(^$I@O\CL"Q@*B DP"6P`V #`"]P('-H;W!P(&$M8NTO0&L?@0JQ M=#0B. (LLOYS$[,DX2="&- 3P" "-]&O`D J< M3'J!L(D!S"'"^9B!B+T F M`S_A(75B">'_-V$DP"FB#; 3T"W@"X M(<DC<F1O!Y!N)P5 -C7_'H B,2OT M*5 C8#9@-/ =G+Y).-$J\1Z0`' VT6$J<_\`P#O ._,>HBLP)X,D$B,R#230 M9R3@,Z)'248O\$I014<@``# ,O(A%/L*P!Z0:B! !4 [D". `_#F9#21.Q!G M<R#P*. T$OYM$U BX3.3)%(@82[4*;'_$\ #$ ,@0/!!-TRB2X G81\%D AQ M/I$D(2*34$]3LE0C8T=%41 8T'$*4/\3P 0@+3$B$$J!.-$E\#.BRSB#//4I M)DY/4B#P0K#_-:,AA"FQ,6$IH#;P+$$-T/=&`0(P-B-M!" EX2C@1%'U,Z)B M`V!W$; %P"10`Z#_-O$UD#>2(10\!$JB(S(K]']1MB#P(W)*2#[((*(DX4/G M`Y%29#/0:7@B,0GP3X#?.J$SX05 -_(J('@$``5 ?29=03>S$;!742W!*^53 M_U'02X!9(S:C`-!$85JA4-#P4D]862#P+<%&@ L@'R!B6ZPD$$1A7B!!0TB^ M12#P$\ %L"!B,Z)D*5 &863#(R-#3TY.171#5%(A+2-@)2(%$&>R:&"?4F4I MX L@<QV</DHO0"0A!Q!"$!/!863W"H43I 0@11O00B$$D!X&V6D@+4%*\1]0 M; <P(/ ^30N *; *A0K ;1-J0%M"P"10+BZ@! `N!:!M"QV<%_$`<Y ```,` M-@```````P`F```````"`?D_`0```!X`````````W*= R,!"$!JTN0@`*R_A M@@$`````````+@```!X`^#\!````%0```%-Y<W1E;2!!9&UI;FES=')A=&]R M``````(!^S\!````'@````````#<IT#(P$(0&K2Y" `K+^&"`0`````````N M````'@#Z/P$````5````4WES=&5M($%D;6EN:7-T<F%T;W(`````0 `',&!! M,=R*J+P!0 `(,(#K]Z>-J+P!`P`--/T_```"`10T`0```! ```!4E*' *7\0 M&Z6'" `K*B47'@`]``$````!``````````L`*0``````"P`C```````"`7\` M`0```#H````\8SU54R5A/5\E<#U55$,E;#U55$,O3U154TE-0R\P,#!$,$,Q 9-4!O='5S<VEM8RYO=&ES+F-O;3X```"Y^CU5 ` end Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA24284; Thu, 14 Aug 1997 02:10:28 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 02:10:28 -0700 (PDT) From: MSF on Squidlist <tech@voerde.globvill.de> Message-Id: <199708140907.LAA25918@brian.voerde.globvill.de> Subject: WS-FTP and Squid To: squid-users@nlanr.net Date: Thu, 14 Aug 1997 11:07:26 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Resent-Message-ID: <"_hWiSt8jLi1.0.ws4.ohiyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/235 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, what options do I have to set to make WS-FTP32 work with Squid? CU, Marcus Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA04036; Thu, 14 Aug 1997 02:33:56 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 02:33:56 -0700 (PDT) Message-Id: <199708140931.CAA03862@nlanr.net> Date: Thu, 14 Aug 1997 11:31:22 +0200 From: "ing. AG van Loo" <eddy@cis.HZeeland.nl> To: squid-users@nlanr.net Subject: Re: Caching violates copyright? Resent-Message-ID: <"fT5sV7JoenC.0.cg1.A2jyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/236 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi List, : > >Thank you kindly for asking. I have a simple question that perhaps you can : > >help me out with. If a page should change at my end, after it has been : > >cached on a proxy server, how is the change reflected in the cached version? : > : > Sounds like he was worried about people not getting the most up-to-date : > sales info. : : This seems to be a pretty common concern from people who don't understand : a lot about web caching. Perhaps a explanatory page could be added to the : Squid web page, going over the miscellany of Expires: headers and : If-Modified-Since requests? : : -Mike Pelletier. And how about some piece of information about -in my opinion- badly configured webservers: Apache has by default: # CacheNegotiatedDocs: By default, Apache sends Pragma: no-cache with each # document that was negotiated on the basis of content. This asks proxy # servers not to cache the document. Uncommenting the following line disables # this behavior, and proxies will be allowed to cache the documents. # CacheNegotiatedDocs Every website that uses an out-of-the-box configuration, like Slackware has this turned off by default.. Personally i feel that this option SHOULD be ON by default ! Regards, Eddy -- Network Management, Unix Support, Surfnet-IC/ICP, Postmaster, Webmaster at Hogeschool Zeeland, Vlissingen / Zeeland Polytechnic, Flushing, Holland. Email: eddy@HZeeland.nl WWW: http://www.HZeeland.nl/~eddy/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id CAA15696; Thu, 14 Aug 1997 02:53:33 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 02:53:33 -0700 (PDT) Date: Thu, 14 Aug 1997 05:42:56 -0400 From: "Armistead, Jason" <ARMISTEJ@oeca.otis.com> Subject: RE: Question about SSL using CONNECT method To: Redfern Ian <RedfernI@logica.com> Cc: "'Squid-Users'" <squid-users@nlanr.net> Message-id: <c=US%a=_%p=UTC%l=UTC/OTUSIMC/000D0C8B@otussimc.otis.com> X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.995.52 Encoding: 46 TEXT Resent-Message-ID: <"MgoxtvN5T26.0.1j.EKjyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/237 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Thanks Ian for the prompt reply. Checking my store.log I see that yes, any access-controlled object only appears as a "RELEASE" entry, and icp.c also indicates that it is indeed non-cacheable. Making GIFs and JPEGs anonymously-readable sounds like it requires a code hack in icp.c which tests not only for authorisation, but for the file extension too. Basically say, "what the heck" to GIF or JPG file extensions. Quite do-able if I had the time. Anyone already created such a hack and care to share the code ? I wish that we weren't so authentication-crazed around my company some times. Just putting the images in a different realm (moving the protected documents down a different directory tree or images moved to another server) would do the trick, wouldn't it ? Anyone care to shed more light on this thorny subject ? Regards Jason Armistead Systems Engineer OEC-Australia, Minto armistej@oeca.otis.com >---------- >From: Redfern Ian[SMTP:RedfernI@logica.com] >Sent: Thursday, 14 August 1997 19:10 >To: Armistead, Jason >Subject: RE: Question about SSL using CONNECT method > >Note that access-controlled objects are not cached either, so you need >to make them anonymously readable. You don't have to make the >directories browsable, though. Then turn <img src="/images/logo.gif"> >into <img src="http:/images/logo.gif"> and you should be OK. This is >easy to automate; alternatively (and more kludgily) you could have >https:/images point to a program that issues redirects (you just avoid >the data transfer hit then, so it's not as good). > >Ian Redfern (redferni@logica.com). > Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id DAA09951; Thu, 14 Aug 1997 03:21:18 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 03:21:18 -0700 (PDT) Message-Id: <m0wyx0I-000cugC@lx01.tietosavo.fi> Comments: Authenticated sender is <jt@pop3> From: "Jussi Torhonen" <jussi.torhonen@tietosavo.fi> Organization: Tietosavo Oy, Kuopio, Finland To: MSF on Squidlist <tech@voerde.globvill.de> Date: Thu, 14 Aug 1997 13:19:40 +0200 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Re: WS-FTP and Squid CC: squid-users@nlanr.net Priority: normal In-reply-to: <199708140907.LAA25918@brian.voerde.globvill.de> X-mailer: Pegasus Mail for Win32 (v2.54) Resent-Message-ID: <"EYsOPIm7czE.0.xr6.Wljyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/238 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On 14 Aug 97 at 11:07, squid-users@nlanr.net wrote: > what options do I have to set to make WS-FTP32 work with Squid? It's not possible at all. But ftp-gw gateway from TIS Internet Firewall Toolkit works fine with WS_FTP. See http://www.tis.com/docs/products/fwtk/ Jussi -*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Jussi Torhonen / Internet: jussi.torhonen@tietosavo.fi Tietosavo Oy / X.400: C=fi, ADMD=elisa, PRMD=tsnet, P.O.Box 1582 / O=tietosavo, S=torhonen, G=jussi FIN-70461 KUOPIO / AX.25: OH7DC@OH7RBA.#KUO.FIN.EU FINLAND / Tel: +358-17-193 231, Fax: +358-17-193 355 -*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id EAA19295; Thu, 14 Aug 1997 04:24:17 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 04:24:17 -0700 (PDT) Message-Id: <199708141118.NAA23372@public.ndh.com> X-Mailer: Calypso Version 2.20.15 Date: Thu, 14 Aug 1997 13:13:01 +0200 From: "Markus Muss" <mm@ndh.net> To: squid-users@nlanr.net Subject: squid(?) problem with apache-agentlog Resent-Message-ID: <"MPeMMWWX7i5.0.yN2.5dkyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/239 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, yesterday I ran into some trouble with a customers apache and my squid. I wanted to enable apache's agentlog by useing the oldfashioned way (mod_agent_log) and not going via custom_log (I will try that today). But when I started the apache and tried to access the pages via my squid it didn't work. Netscape said "connected..." but it couldn't recieve any data. Without using squid everything worked fine (netscape and telnet). The squid-logs didn't contain anything unnormal.... Has anyone any ideas? thanks and ciao, markus Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id FAA12335; Thu, 14 Aug 1997 05:45:54 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 05:45:54 -0700 (PDT) Message-Id: <3.0.3.32.19970814133629.00c61750@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.3 (32) Date: Thu, 14 Aug 1997 13:36:29 +0100 To: Kerry Jones <kjones@aims.gov.au>, squid-users@nlanr.net From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Parents of Parent Proxies In-Reply-To: <3.0.1.32.19970814163716.00886360@email.aims.gov.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"Idp4qdIWHHA.0.q05.Yqlyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/240 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 16:37 14/08/97 +1000, Kerry Jones wrote: >Hi, > >We have configured our Squid Proxy server as a sibling of another Proxy >server and our Parent is currently using another Proxy server as its parent. > >Do we need to configure our Parent's Parent proxies as well as our Parent >proxy or will a request from us to our Parent handle the requests to our >Parent's Parent?? [ I'll call your parent A and your parent's parent B. ] You don't need to, but there are advantages if you did if A goes down. What you could do is put both in as parents in squid.conf, but add a "weight=<number>" to A. That way it prefers A, but will use B if necessary. Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id GAA09326; Thu, 14 Aug 1997 06:49:34 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 06:49:34 -0700 (PDT) Date: Thu, 14 Aug 1997 09:46:29 -0400 (EDT) From: Michael Pelletier <mikep@comshare.com> To: "ing. AG van Loo" <eddy@cis.HZeeland.nl> cc: squid-users@nlanr.net Subject: Re: Caching violates copyright? In-Reply-To: <199708140931.CAA03862@nlanr.net> Message-ID: <Pine.BSI.3.95.970814094502.16605D-100000@valhalla.comshare.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"CkBUtLzT15E.0.WT2.Eomyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/241 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Thu, 14 Aug 1997, ing. AG van Loo wrote: > > Apache has by default: > > # CacheNegotiatedDocs: By default, Apache sends Pragma: no-cache with each > # document that was negotiated on the basis of content. This asks proxy > # servers not to cache the document. Uncommenting the following line disables > # this behavior, and proxies will be allowed to cache the documents. > > Every website that uses an out-of-the-box configuration, like Slackware > has this turned off by default.. > > Personally i feel that this option SHOULD be ON by default ! If it was negotiated, though, doesn't that mean that the content is subject to change based on individual user input? If it's cached, won't that mean that there's a possibility of a user getting someone else's negotiated document? I guess I'm not really clear what they mean by this... -Mike Pelletier. Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA03739; Thu, 14 Aug 1997 07:16:43 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 07:16:43 -0700 (PDT) Date: 14 Aug 1997 15:39:00 +0200 From: ar@andromeda.netaccess.de (Andreas Reeh) To: squid-users@nlanr.net Message-ID: <6bpUSalp4lB@andromeda.netaccess.de> In-Reply-To: <6blRj9w$4lB@andromeda.netaccess.de> Subject: Re: Meaning of "Page faults with physical IO" ? X-Mailer: CrossPoint v3.11 R/C464 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Organization: NetAccess Datendienste GmbH, http://www.netaccess.de/ or info@netaccess.de Resent-Message-ID: <"90vai4pKfl3.0.Eo2.fCnyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/242 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net JLarmour@origin-at.co.uk wrote at 17:33 13 Aug 97 in 3.0.2.32.19970813173301.00c6077c@mailhost.origin-at.co.uk: >>what does "Page faults with physical i/o: 4897" in my Cachemgr "Cache >>Information"-output mean ? >You get a "page fault", when your OS tries to access something in memory >which is actually swapped to disk. The term "page fault", while correct at Fine, that is the same I was thinking, but I wanted to be sure. >Also, this doesn't necessarily mean your squid is swapping by that much. >Most operating systems also implement paging for executables, so that only >However, if the number of faults is unusually high, and getting bigger, what is unusually high ? I asked this because I am wondering a little: The machine and Squid are now running 1 day, 6 hours, and Cachemgr says: Squid Object Cache: Version 1.NOVM.15 Start Time: Wed, 13 Aug 1997 10:55:23 GMT Current Time: Thu, 14 Aug 1997 13:16:00 GMT Connection information for squid: Number of TCP connections: 13480 Number of UDP connections: 0 Connections per hour: 511.7 Select loop called: 556715 times, 170.351 ms avg Cache information for squid: Storage Swap size: 2543 MB Maximum Swap size: 2640 MB Storage LRU Expiration Age: 71.10 days Requests given to unlinkd: 1057 Unused fileno stack count: 56 Resource usage for squid: CPU Time: 1089 seconds (492 user 597 sys) CPU Usage: 1% Maximum Resident Size: 0 KB Page faults with physical i/o: 51826 ^^^^^ Ok, but procinfo says: Memory: Total Used Free Shared Buffers Cached Mem: 63468 62344 1124 22728 14312 14680 Swap: 66540 40 66500 Bootup: Wed Aug 13 09:21:29 1997 Load average: 0.00 0.00 0.00 2/64 25392 user : 0:22:46.54 1.3% page in : 966636 disk 1: 28680r 58203w nice : 0:00:00.00 0.0% page out: 467220 disk 2: 201747r 92831w system: 0:28:03.66 1.6% swap in : 5 idle : 1d 5:03:55.40 97.2% swap out: 21 uptime: 1d 5:54:45.58 context : 2432173 If what procinfo says is true there were only 5 swap in and 21 swap out - so why has squid 51826 page faults with physical i/o ?! Andreas -- Andreas Reeh = A.Reeh@WaD.org (private) or ar@netaccess.de (business) Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id HAA12706; Thu, 14 Aug 1997 07:40:30 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 07:40:30 -0700 (PDT) Subject: Re: Caching violates copyright? To: squid-users@nlanr.net Date: Thu, 14 Aug 1997 15:39:01 +0100 (BST) In-Reply-To: <Pine.BSI.3.95.970814094502.16605D-100000@valhalla.comshare.com> from "Michael Pelletier" at Aug 14, 97 09:46:29 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Message-Id: <E0wz13G-00022c-00@cygnus.csi.cam.ac.uk> From: WWW server manager <webadm@info.cam.ac.uk> Resent-Message-ID: <"b1kHVs9A-I9.0.ui2.gYnyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/243 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Michael Pelletier wrote: > > On Thu, 14 Aug 1997, ing. AG van Loo wrote: > > > > Apache has by default: > > > > # CacheNegotiatedDocs: By default, Apache sends Pragma: no-cache with each > > # document that was negotiated on the basis of content. This asks proxy > > # servers not to cache the document. Uncommenting the following line disables > > # this behavior, and proxies will be allowed to cache the documents. > > > > Every website that uses an out-of-the-box configuration, like Slackware > > has this turned off by default.. > > > > Personally i feel that this option SHOULD be ON by default ! > > If it was negotiated, though, doesn't that mean that the content is > subject to change based on individual user input? If it's cached, won't > that mean that there's a possibility of a user getting someone else's > negotiated document? > That's certainly my (mis-)understanding of what it's about. Documents where what was returned in response to a request is determined by criteria (language, content-type, content "quality" (e.g. image compression loss), etc.) from among multiple documents (or generated dynamically in the form required). The only time it would be sensible to allow caching of such documents is if you (as origin server admin) know that although the documents could differ, the variants are sufficiently similar that it wouldn't really make much difference which version someone received. (In which case, it's not obvious that there would be any point in supporting multiple variants.) Even if a cache server kept track of all the response headers that could be used in content negotiation, I can't see that it would be reasonable in general to return a cached copy unless it could e.g. check (conditional GET) what the origin server would have sent, and confirm that it has a current cached copy of *that* variant. Simply applying variant selection criteria to any variants that happened to be in the cache would not be sufficient, since the cache could not know whether the origin server could offer a vastly better match to the user's requirements than any of the variants that happened by chance to be in the cache. For example, consider a request specifying that the user would prefer the document in Icelandic, but would accept English if necessary, and the only cached copy was in Russian. Returning the Russian version or rejecting the request as impossible to satisfy would both be unreasonable. If the cache also had the English version, returning that would be more reasonable, but still giving the user poor service when the origin server could have provided precisely what had been requested. [The HTTP 1.1 spec has a lot of comments on caching, and I haven't checked what it says about negotiated documents, so it may mandate a different approach...] John Line -- University of Cambridge WWW manager account (usually John Line) Send general WWW-related enquiries to webmaster@ucs.cam.ac.uk Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA07066; Thu, 14 Aug 1997 08:16:19 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 08:16:19 -0700 (PDT) Message-Id: <3.0.3.32.19970814160820.00c6a624@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.3 (32) Date: Thu, 14 Aug 1997 16:08:20 +0100 To: ar@andromeda.netaccess.de (Andreas Reeh), squid-users@nlanr.net From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Meaning of "Page faults with physical IO" ? In-Reply-To: <6bpUSalp4lB@andromeda.netaccess.de> References: <6blRj9w$4lB@andromeda.netaccess.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"4KnwNIlk6hE.0.Ve2.B3oyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/244 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 15:39 14/08/97 +0200, Andreas Reeh wrote: >user : 0:22:46.54 1.3% page in : 966636 disk 1: 28680r 58203w >nice : 0:00:00.00 0.0% page out: 467220 disk 2: 201747r 92831w >system: 0:28:03.66 1.6% swap in : 5 >idle : 1d 5:03:55.40 97.2% swap out: 21 >uptime: 1d 5:54:45.58 context : 2432173 > >If what procinfo says is true there were only 5 swap in and 21 swap out - so >why has squid 51826 page faults with physical i/o ?! Maybe I _slightly_ oversimplified, because OS's these days (not counting MeSsy Windoze) only swap in and out _data_. They page fault when loading executables and whenever malloc'ed regions are actually _used_ (rather than just allocated) (generally). Executable code gets discarded rather than swapped when memory is short, and reloaded directly from the executable. That's why deleting an executable while its running is a hazardous activity. There are other complications that arise with copy-on-write when fork() is called, or if mmap() is used, but these don't affect squid. Upgrading /bin/sh and a shared libc can be fun sometimes! You must be hardly using your swap at all given your statistics. If you type "free", or "top" you should get statistics saying that your swap usage is near zero - at least according to what you've said above. Jonathan L. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA31159; Thu, 14 Aug 1997 08:22:22 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 08:22:22 -0700 (PDT) Message-ID: <33F32355.CE2BCDB4@siemens.nl> Date: Thu, 14 Aug 1997 17:25:09 +0200 From: Gert-Jan Looy <gj@siemens.nl> Reply-To: gj@siemens.nl Organization: Siemens X-Mailer: Mozilla 4.0 [en] (WinNT; I) MIME-Version: 1.0 To: Andreas Reeh <ar@andromeda.netaccess.de> CC: squid-users@nlanr.net Subject: Re: Meaning of "Page faults with physical IO" ? X-Priority: 3 (Normal) References: <6bpUSalp4lB@andromeda.netaccess.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"eYYShYmN0KE.0.N_3.DAoyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/245 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Andreas Reeh wrote: > > JLarmour@origin-at.co.uk wrote at 17:33 13 Aug 97 > in 3.0.2.32.19970813173301.00c6077c@mailhost.origin-at.co.uk: > > >>what does "Page faults with physical i/o: 4897" in my Cachemgr "Cache > >>Information"-output mean ? [...] > Squid Object Cache: Version 1.NOVM.15 > Start Time: Wed, 13 Aug 1997 10:55:23 GMT > Current Time: Thu, 14 Aug 1997 13:16:00 GMT > Page faults with physical i/o: 51826 > ^^^^^ Hasn't this anything to do with the "NOVM" version? My squid (1.1.14) has run for two weeks now, and has a pagefaultcount of only 12 ! Greetings, _________________________________________________________ Gert-Jan Looy Unix SysAdm Siemens/SNI Nederland N.V. E-mail : gj@siemens.nl Phone/Fax : +31 70 333 2982 / 2948 X400 : NL;400NET;SCN;siemens;zom1;p1;looy;gert-jan Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id IAA22899; Thu, 14 Aug 1997 08:37:05 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 08:37:05 -0700 (PDT) Message-ID: <19970814173546.57306@admin.is.co.za> Date: Thu, 14 Aug 1997 17:35:46 +0200 From: Oskar Pearson <oskar@is.co.za> To: Kip DeGraaf <kip@monroe.lib.mi.us> Cc: squid-users@nlanr.net Subject: Re: Caching violates copyright? References: <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> <199708120902.TAA16515@grey.interact.net.au> <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> <3.0.3.32.19970813085013.03784250@monroe.lib.mi.us> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69e In-Reply-To: <3.0.3.32.19970813085013.03784250@monroe.lib.mi.us>; from Kip DeGraaf on Wed, Aug 13, 1997 at 08:50:13AM -0400 Reply-To: oskar@is.co.za Resent-Message-ID: <"AUmgb3dB9AB.0.Me1.qNoyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/246 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi > Okay, wait a second. Pictographics has gotten the message and understands > what a caching proxy really does. Let's not alienate the gentleman in the Ok - I read my mail in batches, so I hadn't received that mail by the time I wrote this. > (Of course, if he hadn't been reasonable, this sounds like an interesting > idea) If each person on this list simply told him that they were caching things, perhaps he would realise that caching is mailstream, and that he should think before he threatens. > At 02:24 PM 8/13/97 +0200, Oskar Pearson wrote: > >How about this. Every person that is using squid or any other cache > >(or even if you are using netscape with the built-in disk cache turned on) > >send him a (not 5000...) copy of this message with the appropriate stuff > >changed. I just did. Let him know that he should get a clue before > >he goes crazy. > > > >----------- > >To: webmaster@pictographics.com > > > >We (cache.is.co.za) also cache your web pages. If you do not wish them to > >be saved here feel free to sue us. If there are too many of us, perhaps you > >would think about trying to use the STANDARDS, which our software obeys, > >so that we don't cache pages. It's a little akin to putting a digital > >'Copyright' signature on a page... but something that machines can read. > > > >ftp://ftp.isi.edu/in-notes/rfc/rfc2068.txt > >ftp.isi.edu:/in-notes/rfc/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA16259; Thu, 14 Aug 1997 09:04:38 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 09:04:38 -0700 (PDT) Message-ID: <19970814180306.41134@admin.is.co.za> Date: Thu, 14 Aug 1997 18:03:06 +0200 From: Oskar Pearson <oskar@is.co.za> To: Duane Wessels <wessels@nlanr.net> Cc: squid-users@nlanr.net Subject: Re: Thinking about ICP References: <199708140248.TAA21796@nlanr.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.69e In-Reply-To: <199708140248.TAA21796@nlanr.net>; from Duane Wessels on Wed, Aug 13, 1997 at 07:48:25PM -0700 Reply-To: oskar@is.co.za Resent-Message-ID: <"4_rX8Eovhe5.0.9s7.Mnoyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/247 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi > >We currently have this problem at a National Exchange Point where we have > >many siblings querying each other over multiple WAN's. > > > >The idea I put forward was to "cluster" the siblings in groups of 8 like so: > > > > > > [ 8 cluser sibling ] [ 8 cluster sibling ] > > x x > > x x > > x x > > [ ICP relay ] > > x x > > x x > > x x > > [ 8 cluser sibling ] [ 8 cluster sibling ] > > > >All siblings in a cluster would also include the ICP relay system as a sibling. > >The ICP relay system would fire off UDP_REQUESTS to all other siblings in > >the "other" clusters. > >The ICP relay would relay back to the original requestor a single MISS or a HIT with the real proxies h > ostname so the object can be fetched. > >What this elimintates is each sibling having to reply to the requestor with > >a MISS. > > This is a really interesting idea, similar to what others have proposed > or asked for. There is only one thing preventing this from being I was one of them that asked for this... > really easy to implement: Squid and ICP don't have any concept of > third-party caches. At one point there was another cache package > (mirror-image?) which was going to implment third-party "redirections" > with ICP. There are a couple of ways the ICP relay could be > implemented with some modifications/additions to ICP. If anyone is > seriously interested in building one of these, please let me know and > we can work it all out. I wouldn't mind working on this (limited programming experience, I am afraid). Here is our scenario: We (currently) have 3 large caches, when we configure a client cache to point to us as a parent we generally set them up with the 'default no-query' options, so that they hit us for an object. Squid chooses one of the caches as it's default parent (we have one A record pointing to multiple IP addresses). Since it does a TCP request to this cache (say cache1, for example) cache1 will query cache2 and cache3 on it's behalf. This means that we get effective use of our disk cache. If they aren't setup to point to us as a parent, however (due to all sorts of reasons that enforce this), we have to configure them as follows: cache_host sibling cache1 ... cache_host sibling cache2 ... cache_host sibling cache3 ... Now this means that they have 3 queries coming up their line and 3 going down... and if there is one of these packets lost coz they are flooding their line the cache slooows down. When I spoke to you (a while ago) you said: --------- HIT. Then you could use the 'redirect' opcode to make the querying cache go to the right place. --------- 2. ICP Opcodes The following table shows currently defined ICP opcodes: Value Name ----- ----------------- 0 ICP_OP_INVALID 1 ICP_OP_QUERY 2 ICP_OP_HIT 3 ICP_OP_MISS 4 ICP_OP_ERR 5-9 UNUSED 10 ICP_OP_SECHO 11 ICP_OP_DECHO 12-20 UNUSED 21 ICP_OP_MISS_NOFETCH 22 ICP_OP_DENIED 23 ICP_OP_HIT_OBJ I can't see one... if there was one it would be fairly easy (relative to the whole of squid ;') to write a 'concentrator', so they point to cache-sibling.is.co.za, which does a query of each and every sibling and sends a reply, with a 'redirect to cache1.is.co.za'.... there will probably be security implications in handling this though. Oskar Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA08117; Thu, 14 Aug 1997 09:16:45 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 09:16:45 -0700 (PDT) Date: 14 Aug 1997 18:04:00 +0200 From: ar@andromeda.netaccess.de (Andreas Reeh) To: squid-users@nlanr.net Message-ID: <6bpUfwMp4lB@andromeda.netaccess.de> Subject: Speedup by turning of debugging ? X-Mailer: CrossPoint v3.11 R/C464 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Organization: NetAccess Datendienste GmbH, http://www.netaccess.de/ or info@netaccess.de Resent-Message-ID: <"YN9VYf30D-4.0.tv7.xyoyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/248 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi Squiders, in http://cache.is.co.za/squid/opt/performance.html it is written that turning up debugging would result in a 10-15% fast select loop. I have some questions about this: 1) Where/how can I disable debugging ? 2) What exactely happens when debugging is tuerned of, has this any influence on a normal running cache ? I mean what negative things could happen when I turn debugging of - are errors/Problems still logged in cache.log ? (what is it than I turn off and don't have any more? 3) How are your experiences whith this ? 4) Are there any other usefull ways for tuning Squid or the machine squid is running on, not mentioned on the above URL ? (machines are all Pentiums 90-166, 64 MB, SCSII, running Linux 2.0.30 with fd_patch and Squid 1.NOVM.15) And, a little offtoppic: I am searching tips for tuneing an other Linux machine (2xP166 SMP, 128 MB) with with inn as news- and uucp server, too. Perhaps anybody knows some good resources? Many TIA, Andreas -- Andreas Reeh = A.Reeh@WaD.org (private) or ar@netaccess.de (business) Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id JAA12298; Thu, 14 Aug 1997 09:51:11 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 09:51:11 -0700 (PDT) Message-Id: <3.0.3.32.19970814173923.00c6b53c@mailhost.origin-at.co.uk> X-Sender: jlarmour@mailhost.origin-at.co.uk X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.3 (32) Date: Thu, 14 Aug 1997 17:39:23 +0100 To: gj@siemens.nl, Andreas Reeh <ar@andromeda.netaccess.de> From: Jonathan Larmour <JLarmour@origin-at.co.uk> Subject: Re: Meaning of "Page faults with physical IO" ? Cc: squid-users@nlanr.net In-Reply-To: <33F32355.CE2BCDB4@siemens.nl> References: <6bpUSalp4lB@andromeda.netaccess.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"8kGcdI4TiE.0.nN3.4Ppyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/249 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net At 17:25 14/08/97 +0200, Gert-Jan Looy wrote: >Andreas Reeh wrote: >> Page faults with physical i/o: 51826 >> ^^^^^ > >Hasn't this anything to do with the "NOVM" version? >My squid (1.1.14) has run for two weeks now, and has a pagefaultcount of >only 12 ! You'll certainly get different results from the NOVM as you won't get page faults from malloc'ed memory. But in your case, I imagine that its just the way your OS counts page faults. Yours might only change this stat if there is a genuine swap out and in, rather than the other murky things to do with demand-loading executables and malloc'ed memory that I've been talking about. What OS is it? Jonathan L. P.S. For those interested, what happens is that malloc() allocates memory in blocks, and when all the blocks are full up, calls sbrk() which calls mmap(). The result is that in a sense you only reserve extra memory in the page tables, but you don't _actually_ allocate it. Its only when the memory is written to, that it is actually allocated a real chunk of (virtual) memory. Some (most?) OS's make the program seg fault if you try to read from a page of unallocated virtual memory even though it is in your page tables, as this normally indicates you're reading from uninitialised memory. Origin UK,323 Cambridge Science Park,Cambridge,England. Tel: +44(1223)423355 -------[ It is impossible to enjoy idling thoroughly unless one has ]------- ----------------[ plenty of work to do - Jerome K. Jerome ]----------------- Help fight spam! http://spam.abuse.net/ These opinions are all my own fault Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id KAA11272; Thu, 14 Aug 1997 10:07:55 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 10:07:55 -0700 (PDT) Date: Fri, 15 Aug 1997 02:57:41 +1000 (EST) From: Peter Marelas <maral@phase-one.com.au> To: Oskar Pearson <oskar@is.co.za> cc: Duane Wessels <wessels@nlanr.net>, squid-users@nlanr.net Subject: Re: Thinking about ICP In-Reply-To: <19970814180306.41134@admin.is.co.za> Message-ID: <Pine.GSO.3.96.970815025053.25542B-100000@jazzy.phase-one.com.au> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"casZHZRDov2.0.cK2.pipyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/250 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net On Thu, 14 Aug 1997, Oskar Pearson wrote: > Hi > > > >We currently have this problem at a National Exchange Point where we have > > >many siblings querying each other over multiple WAN's. > > > > > >The idea I put forward was to "cluster" the siblings in groups of 8 like so: > > > > > > > > > [ 8 cluser sibling ] [ 8 cluster sibling ] > > > x x > > > x x > > > x x > > > [ ICP relay ] > > > x x > > > x x > > > x x > > > [ 8 cluser sibling ] [ 8 cluster sibling ] > > > > > >All siblings in a cluster would also include the ICP relay system as a sibling. > > >The ICP relay system would fire off UDP_REQUESTS to all other siblings in > > >the "other" clusters. > > >The ICP relay would relay back to the original requestor a single MISS or a HIT with the real proxies h > > ostname so the object can be fetched. > > >What this elimintates is each sibling having to reply to the requestor with > > >a MISS. > > > > This is a really interesting idea, similar to what others have proposed > > or asked for. There is only one thing preventing this from being > I was one of them that asked for this... > > > really easy to implement: Squid and ICP don't have any concept of > > third-party caches. At one point there was another cache package > > (mirror-image?) which was going to implment third-party "redirections" > > with ICP. There are a couple of ways the ICP relay could be > > implemented with some modifications/additions to ICP. If anyone is > > seriously interested in building one of these, please let me know and > > we can work it all out. > > I wouldn't mind working on this (limited programming experience, I am > afraid). Here is our scenario: > > We (currently) have 3 large caches, when we configure a client cache > to point to us as a parent we generally set them up with the > 'default no-query' options, so that they hit us for an object. Squid chooses > one of the caches as it's default parent (we have one A record pointing to > multiple IP addresses). Since it does a TCP request to this cache (say > cache1, for example) cache1 will query cache2 and cache3 on it's behalf. > This means that we get effective use of our disk cache. > > If they aren't setup to point to us as a parent, however (due to all sorts > of reasons that enforce this), we have to configure them as follows: > cache_host sibling cache1 ... > cache_host sibling cache2 ... > cache_host sibling cache3 ... > > Now this means that they have 3 queries coming up their line and 3 going > down... and if there is one of these packets lost coz they are flooding > their line the cache slooows down. > > When I spoke to you (a while ago) you said: > --------- > HIT. Then you could use the 'redirect' opcode to make the querying > cache go to the right place. > --------- > > 2. ICP Opcodes > The following table shows currently defined ICP opcodes: > > Value Name > ----- ----------------- > 0 ICP_OP_INVALID > 1 ICP_OP_QUERY > 2 ICP_OP_HIT > 3 ICP_OP_MISS > 4 ICP_OP_ERR > 5-9 UNUSED > 10 ICP_OP_SECHO > 11 ICP_OP_DECHO > 12-20 UNUSED > 21 ICP_OP_MISS_NOFETCH > 22 ICP_OP_DENIED > 23 ICP_OP_HIT_OBJ > > I can't see one... if there was one it would be fairly easy (relative to > the whole of squid ;') to write a 'concentrator', so they point to > cache-sibling.is.co.za, which does a query of each and every sibling > and sends a reply, with a 'redirect to cache1.is.co.za'.... there will > probably be security implications in handling this though. > Since your client is treating your neighbours as siblings, you could probably easily mimic want you want by configuring a cache (that you manage) to query these 3 siblings for your clients cache. Your client would use this "hacked squid" cache as a parent, but you would hack it so it returns ICP_OP_MISS_NOFETCH when it has not found the object in its own cache and that of the 3 siblings it queries. Note: I am assuming here that the hacked parent will query its siblings for the object before returning a ICP_OP_MISS_NOFETCH. A ICP_OP_MISS_NOFETCH reply effectively tells the originating squid "Yes you have requested an object, but im busy reloading my log, so im not going to fetch the object for you.. Get it yourself". Regards Peter Marelas -- Phase One Interactive - Sun Solaris/Unix/Networking Consultant P.O Box 549, Templestowe 3106 Melbourne, Australia URL: http://www.phase-one.com.au/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id KAA24449; Thu, 14 Aug 1997 10:32:47 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 10:32:47 -0700 (PDT) Sender: joydon@victory.soonet.ca Message-ID: <33F34297.30145C2D@soonet.ca> Date: Thu, 14 Aug 1997 13:38:32 -0400 From: Don Joy <joydon@soonet.ca> Reply-To: joydon@soonet.ca Organization: SooNet Corp X-Mailer: Mozilla 4.02b7 [en] (X11; I; BSD/OS 3.0 i386) MIME-Version: 1.0 To: oskar@is.co.za CC: squid-users@nlanr.net Subject: Re: Caching violates copyright? References: <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> <199708120902.TAA16515@grey.interact.net.au> <Pine.LNX.3.96.970812211727.17462B-100000@linkou.trace.com.tw> <3.0.3.32.19970813085013.03784250@monroe.lib.mi.us> <19970814173546.57306@admin.is.co.za> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Resent-Message-ID: <"FFWXFrLOe-7.0.7t6.64qyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/252 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Well, this is rather interesting. I haven't been following this thread really closely, but i happened to notice the domain in question here. As it turns out, this individual is one of our isdn customers(small world). I guess he doesn't realize that we too run a proxy server, and guess what, his pages are cached here too:). Oskar Pearson wrote: > Hi > > > Okay, wait a second. Pictographics has gotten the message and understands > > what a caching proxy really does. Let's not alienate the gentleman in the > Ok - I read my mail in batches, so I hadn't received that mail by the time > I wrote this. > > > (Of course, if he hadn't been reasonable, this sounds like an interesting > > idea) > If each person on this list simply told him that they were caching > things, perhaps he would realise that caching is mailstream, and that > he should think before he threatens. > > > At 02:24 PM 8/13/97 +0200, Oskar Pearson wrote: > > >How about this. Every person that is using squid or any other cache > > >(or even if you are using netscape with the built-in disk cache turned on) > > >send him a (not 5000...) copy of this message with the appropriate stuff > > >changed. I just did. Let him know that he should get a clue before > > >he goes crazy. > > > > > >----------- > > >To: webmaster@pictographics.com > > > > > >We (cache.is.co.za) also cache your web pages. If you do not wish them to > > >be saved here feel free to sue us. If there are too many of us, perhaps you > > >would think about trying to use the STANDARDS, which our software obeys, > > >so that we don't cache pages. It's a little akin to putting a digital > > >'Copyright' signature on a page... but something that machines can read. > > > > > >ftp://ftp.isi.edu/in-notes/rfc/rfc2068.txt > > >ftp.isi.edu:/in-notes/rfc/ Received: (from list@localhost) by nlanr.net (8.8.6/8.8.6) id KAA09165; Thu, 14 Aug 1997 10:32:48 -0700 (PDT) Resent-Date: Thu, 14 Aug 1997 10:32:48 -0700 (PDT) To: squid-users@nlanr.net Path: not-for-mail From: Christophe Zwecker <doc@rrz.uni-hamburg.de> Newsgroups: Local.squid Subject: how never expire ? Date: Thu, 14 Aug 1997 17:27:12 +0200 Lines: 15 Message-Id: <33F323D0.7DD134D6@rrz.uni-hamburg.de> Nntp-Posting-Host: localhost Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 3.01 (X11; I; Linux 2.0.30 i586) Resent-Message-ID: <"fsDDFSzAWC6.0.4u5.M3qyp"@oceana.nlanr.net> Resent-From: squid-users@nlanr.net X-Mailing-List: <squid-users@nlanr.net> archive/latest/251 X-Loop: squid-users@nlanr.net Precedence: list Resent-Sender: squid-users-request@nlanr.net Hi, I don'T want the cache to expire ever - what Di I have to put in squid.conf ? bye -- Christophe ---------------------------------------------------------------------- Christophe Zwecker email: doc@rrz.uni-hamburg.de Hamburg, Germany fax: 49 40 22715433 ---------------------------------------------------------------------- In newsgroups please reply also by email - thanksReceived on Thu Aug 07 1997 - 13:24:32 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:36:46 MST